Spaces:
Running
Running
Update single-image results, add model link url
Browse files- app.py +17 -16
- static/css/{core_single.css → single_image.css} +12 -12
- static/eval_results/Core_SI/all_summary.json +0 -227
- static/eval_results/Default/all_model_keywords_stats.json +160 -160
- static/eval_results/Default/all_summary.json +68 -134
- static/eval_results/{Core_SI → SI}/all_model_keywords_stats.json +0 -0
- static/eval_results/SI/all_summary.json +471 -0
- utils.py +112 -17
app.py
CHANGED
|
@@ -1,5 +1,5 @@
|
|
| 1 |
import gradio as gr
|
| 2 |
-
from utils import DefaultDataLoader,
|
| 3 |
import os
|
| 4 |
from constants import *
|
| 5 |
|
|
@@ -9,19 +9,19 @@ current_dir = os.path.dirname(os.path.abspath(__file__))
|
|
| 9 |
# Construct paths to CSS files
|
| 10 |
base_css_file = os.path.join(current_dir, "static", "css", "style.css")
|
| 11 |
default_css_file = os.path.join(current_dir, "static", "css", "default.css")
|
| 12 |
-
|
| 13 |
|
| 14 |
# Read CSS files
|
| 15 |
with open(base_css_file, "r") as f:
|
| 16 |
base_css = f.read()
|
| 17 |
with open(default_css_file, "r") as f:
|
| 18 |
default_css = f.read()
|
| 19 |
-
with open(
|
| 20 |
-
|
| 21 |
|
| 22 |
# Initialize data loaders
|
| 23 |
default_loader = DefaultDataLoader()
|
| 24 |
-
|
| 25 |
|
| 26 |
with gr.Blocks() as block:
|
| 27 |
# Add a style element that we'll update
|
|
@@ -49,14 +49,14 @@ with gr.Blocks() as block:
|
|
| 49 |
|
| 50 |
with gr.Row():
|
| 51 |
table_selector = gr.Radio(
|
| 52 |
-
choices=["Default", "
|
| 53 |
-
label="Select table to display",
|
| 54 |
value="Default"
|
| 55 |
)
|
| 56 |
|
| 57 |
# Define different captions for each table
|
| 58 |
default_caption = "**Table 1: MEGA-Bench full results.** The number in the parentheses is the number of tasks of each keyword. <br> The Core set contains $N_{\\text{core}} = 440$ tasks evaluated by rule-based metrics, and the Open-ended set contains $N_{\\text{open}} = 65$ tasks evaluated by a VLM judge (we use GPT-4o-0806). <br> $\\text{Overall} \\ = \\ \\frac{\\max(\\text{Core w/o CoT}, \\ \\text{Core w/ CoT}) \\ \\cdot \\ N_{\\text{core}} \\ + \\ \\text{Open-ended} \\ \\cdot \\ N_{\\text{open}}}{N_{\\text{core}} \\ + \\ N_{\\text{open}}}$"
|
| 59 |
-
|
| 60 |
|
| 61 |
caption_component = gr.Markdown(
|
| 62 |
value=default_caption,
|
|
@@ -80,10 +80,10 @@ with gr.Blocks() as block:
|
|
| 80 |
data_component = gr.Dataframe(
|
| 81 |
value=initial_data,
|
| 82 |
headers=initial_headers,
|
| 83 |
-
datatype=["
|
| 84 |
interactive=False,
|
| 85 |
elem_classes="custom-dataframe",
|
| 86 |
-
max_height=
|
| 87 |
)
|
| 88 |
|
| 89 |
def update_table_and_caption(table_type, super_group, model_group):
|
|
@@ -91,23 +91,24 @@ with gr.Blocks() as block:
|
|
| 91 |
headers, data = default_loader.get_leaderboard_data(super_group, model_group)
|
| 92 |
caption = default_caption
|
| 93 |
current_css = f"{base_css}\n{default_css}"
|
| 94 |
-
else: #
|
| 95 |
-
headers, data =
|
| 96 |
-
caption =
|
| 97 |
-
current_css = f"{base_css}\n{
|
| 98 |
|
| 99 |
return [
|
| 100 |
gr.Dataframe(
|
| 101 |
value=data,
|
| 102 |
headers=headers,
|
| 103 |
-
datatype=["
|
|
|
|
| 104 |
),
|
| 105 |
caption,
|
| 106 |
f"<style>{current_css}</style>"
|
| 107 |
]
|
| 108 |
|
| 109 |
def update_selectors(table_type):
|
| 110 |
-
loader = default_loader if table_type == "Default" else
|
| 111 |
return [
|
| 112 |
gr.Radio(choices=list(loader.SUPER_GROUPS.keys())),
|
| 113 |
gr.Radio(choices=list(loader.MODEL_GROUPS.keys()))
|
|
|
|
| 1 |
import gradio as gr
|
| 2 |
+
from utils import DefaultDataLoader, SingleImageDataLoader
|
| 3 |
import os
|
| 4 |
from constants import *
|
| 5 |
|
|
|
|
| 9 |
# Construct paths to CSS files
|
| 10 |
base_css_file = os.path.join(current_dir, "static", "css", "style.css")
|
| 11 |
default_css_file = os.path.join(current_dir, "static", "css", "default.css")
|
| 12 |
+
si_css_file = os.path.join(current_dir, "static", "css", "single_image.css")
|
| 13 |
|
| 14 |
# Read CSS files
|
| 15 |
with open(base_css_file, "r") as f:
|
| 16 |
base_css = f.read()
|
| 17 |
with open(default_css_file, "r") as f:
|
| 18 |
default_css = f.read()
|
| 19 |
+
with open(si_css_file, "r") as f:
|
| 20 |
+
si_css = f.read()
|
| 21 |
|
| 22 |
# Initialize data loaders
|
| 23 |
default_loader = DefaultDataLoader()
|
| 24 |
+
si_loader = SingleImageDataLoader()
|
| 25 |
|
| 26 |
with gr.Blocks() as block:
|
| 27 |
# Add a style element that we'll update
|
|
|
|
| 49 |
|
| 50 |
with gr.Row():
|
| 51 |
table_selector = gr.Radio(
|
| 52 |
+
choices=["Default", "Single Image"],
|
| 53 |
+
label="Select table to display. Default: all MEGA-Bench tasks; Single Image: single-image tasks only.",
|
| 54 |
value="Default"
|
| 55 |
)
|
| 56 |
|
| 57 |
# Define different captions for each table
|
| 58 |
default_caption = "**Table 1: MEGA-Bench full results.** The number in the parentheses is the number of tasks of each keyword. <br> The Core set contains $N_{\\text{core}} = 440$ tasks evaluated by rule-based metrics, and the Open-ended set contains $N_{\\text{open}} = 65$ tasks evaluated by a VLM judge (we use GPT-4o-0806). <br> $\\text{Overall} \\ = \\ \\frac{\\max(\\text{Core w/o CoT}, \\ \\text{Core w/ CoT}) \\ \\cdot \\ N_{\\text{core}} \\ + \\ \\text{Open-ended} \\ \\cdot \\ N_{\\text{open}}}{N_{\\text{core}} \\ + \\ N_{\\text{open}}}$"
|
| 59 |
+
single_image_caption = "**Table 2: MEGA-Bench Single-image setting results.** The number in the parentheses is the number of tasks in each keyword. <br> This subset contains 273 single-image tasks from the Core set and 42 single-image tasks from the Open-ended set. For open-source models, we drop the image input in the 1-shot demonstration example so that the entire query contains a single image only. <br> Compared to the default table, some models with only single-image support are added."
|
| 60 |
|
| 61 |
caption_component = gr.Markdown(
|
| 62 |
value=default_caption,
|
|
|
|
| 80 |
data_component = gr.Dataframe(
|
| 81 |
value=initial_data,
|
| 82 |
headers=initial_headers,
|
| 83 |
+
datatype=["html"] + ["number"] * (len(initial_headers) - 1),
|
| 84 |
interactive=False,
|
| 85 |
elem_classes="custom-dataframe",
|
| 86 |
+
max_height=2400,
|
| 87 |
)
|
| 88 |
|
| 89 |
def update_table_and_caption(table_type, super_group, model_group):
|
|
|
|
| 91 |
headers, data = default_loader.get_leaderboard_data(super_group, model_group)
|
| 92 |
caption = default_caption
|
| 93 |
current_css = f"{base_css}\n{default_css}"
|
| 94 |
+
else: # Single-image
|
| 95 |
+
headers, data = si_loader.get_leaderboard_data(super_group, model_group)
|
| 96 |
+
caption = single_image_caption
|
| 97 |
+
current_css = f"{base_css}\n{si_css}"
|
| 98 |
|
| 99 |
return [
|
| 100 |
gr.Dataframe(
|
| 101 |
value=data,
|
| 102 |
headers=headers,
|
| 103 |
+
datatype=["html"] + ["number"] * (len(headers) - 1),
|
| 104 |
+
interactive=False,
|
| 105 |
),
|
| 106 |
caption,
|
| 107 |
f"<style>{current_css}</style>"
|
| 108 |
]
|
| 109 |
|
| 110 |
def update_selectors(table_type):
|
| 111 |
+
loader = default_loader if table_type == "Default" else si_loader
|
| 112 |
return [
|
| 113 |
gr.Radio(choices=list(loader.SUPER_GROUPS.keys())),
|
| 114 |
gr.Radio(choices=list(loader.MODEL_GROUPS.keys()))
|
static/css/{core_single.css → single_image.css}
RENAMED
|
@@ -1,18 +1,18 @@
|
|
| 1 |
-
.custom-dataframe thead th:nth-child(-n+
|
| 2 |
-
.custom-dataframe tbody td:nth-child(-n+
|
| 3 |
background-color: var(--global-column-background) !important;
|
| 4 |
}
|
| 5 |
|
| 6 |
-
.custom-dataframe thead th:nth-child(n+
|
| 7 |
-
.custom-dataframe tbody td:nth-child(n+
|
| 8 |
background-color: var(--dimension-column-background) !important;
|
| 9 |
}
|
| 10 |
|
| 11 |
-
.custom-dataframe tbody tr:nth-child(even) td:nth-child(-n+
|
| 12 |
background-color: var(--row-even-global) !important;
|
| 13 |
}
|
| 14 |
|
| 15 |
-
.custom-dataframe tbody tr:nth-child(even) td:nth-child(n+
|
| 16 |
background-color: var(--row-even-dimension) !important;
|
| 17 |
}
|
| 18 |
|
|
@@ -33,21 +33,21 @@
|
|
| 33 |
color: var(--text-color) !important;
|
| 34 |
}
|
| 35 |
|
| 36 |
-
.custom-dataframe thead th:nth-child(-n+
|
| 37 |
-
.custom-dataframe tbody td:nth-child(-n+
|
| 38 |
background-color: var(--global-column-background) !important;
|
| 39 |
}
|
| 40 |
|
| 41 |
-
.custom-dataframe thead th:nth-child(n+
|
| 42 |
-
.custom-dataframe tbody td:nth-child(n+
|
| 43 |
background-color: var(--dimension-column-background) !important;
|
| 44 |
}
|
| 45 |
|
| 46 |
-
.custom-dataframe tbody tr:nth-child(even) td:nth-child(-n+
|
| 47 |
background-color: var(--row-even-global) !important;
|
| 48 |
}
|
| 49 |
|
| 50 |
-
.custom-dataframe tbody tr:nth-child(even) td:nth-child(n+
|
| 51 |
background-color: var(--row-even-dimension) !important;
|
| 52 |
}
|
| 53 |
|
|
|
|
| 1 |
+
.custom-dataframe thead th:nth-child(-n+4),
|
| 2 |
+
.custom-dataframe tbody td:nth-child(-n+4) {
|
| 3 |
background-color: var(--global-column-background) !important;
|
| 4 |
}
|
| 5 |
|
| 6 |
+
.custom-dataframe thead th:nth-child(n+5),
|
| 7 |
+
.custom-dataframe tbody td:nth-child(n+5) {
|
| 8 |
background-color: var(--dimension-column-background) !important;
|
| 9 |
}
|
| 10 |
|
| 11 |
+
.custom-dataframe tbody tr:nth-child(even) td:nth-child(-n+4) {
|
| 12 |
background-color: var(--row-even-global) !important;
|
| 13 |
}
|
| 14 |
|
| 15 |
+
.custom-dataframe tbody tr:nth-child(even) td:nth-child(n+5) {
|
| 16 |
background-color: var(--row-even-dimension) !important;
|
| 17 |
}
|
| 18 |
|
|
|
|
| 33 |
color: var(--text-color) !important;
|
| 34 |
}
|
| 35 |
|
| 36 |
+
.custom-dataframe thead th:nth-child(-n+4),
|
| 37 |
+
.custom-dataframe tbody td:nth-child(-n+4) {
|
| 38 |
background-color: var(--global-column-background) !important;
|
| 39 |
}
|
| 40 |
|
| 41 |
+
.custom-dataframe thead th:nth-child(n+5),
|
| 42 |
+
.custom-dataframe tbody td:nth-child(n+5) {
|
| 43 |
background-color: var(--dimension-column-background) !important;
|
| 44 |
}
|
| 45 |
|
| 46 |
+
.custom-dataframe tbody tr:nth-child(even) td:nth-child(-n+4) {
|
| 47 |
background-color: var(--row-even-global) !important;
|
| 48 |
}
|
| 49 |
|
| 50 |
+
.custom-dataframe tbody tr:nth-child(even) td:nth-child(n+5) {
|
| 51 |
background-color: var(--row-even-dimension) !important;
|
| 52 |
}
|
| 53 |
|
static/eval_results/Core_SI/all_summary.json
DELETED
|
@@ -1,227 +0,0 @@
|
|
| 1 |
-
{
|
| 2 |
-
"Aquila_VL_2B": {
|
| 3 |
-
"num_eval_tasks": 273,
|
| 4 |
-
"num_eval_samples": 4116,
|
| 5 |
-
"num_not_eval_samples": 0,
|
| 6 |
-
"num_total_samples": 4377,
|
| 7 |
-
"macro_mean_score": 0.20770364903712493,
|
| 8 |
-
"micro_mean_score": 0.20333142638522636,
|
| 9 |
-
"missing_tasks": []
|
| 10 |
-
},
|
| 11 |
-
"Aria": {
|
| 12 |
-
"num_eval_tasks": 273,
|
| 13 |
-
"num_eval_samples": 4116,
|
| 14 |
-
"num_not_eval_samples": 0,
|
| 15 |
-
"num_total_samples": 4377,
|
| 16 |
-
"macro_mean_score": 0.3178882776147889,
|
| 17 |
-
"micro_mean_score": 0.3101511832828904,
|
| 18 |
-
"missing_tasks": []
|
| 19 |
-
},
|
| 20 |
-
"Claude_3.5": {
|
| 21 |
-
"num_eval_tasks": 273,
|
| 22 |
-
"num_eval_samples": 4116,
|
| 23 |
-
"num_not_eval_samples": 0,
|
| 24 |
-
"num_total_samples": 4116,
|
| 25 |
-
"macro_mean_score": 0.520276385877485,
|
| 26 |
-
"micro_mean_score": 0.520276385877485
|
| 27 |
-
},
|
| 28 |
-
"Claude_3.5_new": {
|
| 29 |
-
"num_eval_tasks": 273,
|
| 30 |
-
"num_eval_samples": 4116,
|
| 31 |
-
"num_not_eval_samples": 0,
|
| 32 |
-
"num_total_samples": 4116,
|
| 33 |
-
"macro_mean_score": 0.5462752278980763,
|
| 34 |
-
"micro_mean_score": 0.5462752278980763
|
| 35 |
-
},
|
| 36 |
-
"GPT_4o": {
|
| 37 |
-
"num_eval_tasks": 273,
|
| 38 |
-
"num_eval_samples": 4116,
|
| 39 |
-
"num_not_eval_samples": 0,
|
| 40 |
-
"num_total_samples": 4116,
|
| 41 |
-
"macro_mean_score": 0.5529953662872719,
|
| 42 |
-
"micro_mean_score": 0.5529953662872719
|
| 43 |
-
},
|
| 44 |
-
"GPT_4o_mini": {
|
| 45 |
-
"num_eval_tasks": 273,
|
| 46 |
-
"num_eval_samples": 4116,
|
| 47 |
-
"num_not_eval_samples": 0,
|
| 48 |
-
"num_total_samples": 4116,
|
| 49 |
-
"macro_mean_score": 0.44285970964797233,
|
| 50 |
-
"micro_mean_score": 0.44285970964797233
|
| 51 |
-
},
|
| 52 |
-
"Gemini_1.5_flash_002": {
|
| 53 |
-
"num_eval_tasks": 273,
|
| 54 |
-
"num_eval_samples": 4116,
|
| 55 |
-
"num_not_eval_samples": 0,
|
| 56 |
-
"num_total_samples": 4116,
|
| 57 |
-
"macro_mean_score": 0.42188460865574384,
|
| 58 |
-
"micro_mean_score": 0.42188460865574384
|
| 59 |
-
},
|
| 60 |
-
"Gemini_1.5_pro_002": {
|
| 61 |
-
"num_eval_tasks": 273,
|
| 62 |
-
"num_eval_samples": 4116,
|
| 63 |
-
"num_not_eval_samples": 0,
|
| 64 |
-
"num_total_samples": 4116,
|
| 65 |
-
"macro_mean_score": 0.4914311038229404,
|
| 66 |
-
"micro_mean_score": 0.4914311038229404
|
| 67 |
-
},
|
| 68 |
-
"Idefics3": {
|
| 69 |
-
"num_eval_tasks": 273,
|
| 70 |
-
"num_eval_samples": 4116,
|
| 71 |
-
"num_not_eval_samples": 0,
|
| 72 |
-
"num_total_samples": 4377,
|
| 73 |
-
"macro_mean_score": 0.08941182847569326,
|
| 74 |
-
"micro_mean_score": 0.08779475233900695,
|
| 75 |
-
"missing_tasks": []
|
| 76 |
-
},
|
| 77 |
-
"InternVL2_2B": {
|
| 78 |
-
"num_eval_tasks": 273,
|
| 79 |
-
"num_eval_samples": 4116,
|
| 80 |
-
"num_not_eval_samples": 0,
|
| 81 |
-
"num_total_samples": 4377,
|
| 82 |
-
"macro_mean_score": 0.12069001041308772,
|
| 83 |
-
"micro_mean_score": 0.11842605219090299,
|
| 84 |
-
"missing_tasks": []
|
| 85 |
-
},
|
| 86 |
-
"InternVL2_76B": {
|
| 87 |
-
"num_eval_tasks": 273,
|
| 88 |
-
"num_eval_samples": 4116,
|
| 89 |
-
"num_not_eval_samples": 0,
|
| 90 |
-
"num_total_samples": 4377,
|
| 91 |
-
"macro_mean_score": 0.3998616568018755,
|
| 92 |
-
"micro_mean_score": 0.39149064302628933,
|
| 93 |
-
"missing_tasks": []
|
| 94 |
-
},
|
| 95 |
-
"InternVL2_8B": {
|
| 96 |
-
"num_eval_tasks": 273,
|
| 97 |
-
"num_eval_samples": 4116,
|
| 98 |
-
"num_not_eval_samples": 0,
|
| 99 |
-
"num_total_samples": 4377,
|
| 100 |
-
"macro_mean_score": 0.27650612401825575,
|
| 101 |
-
"micro_mean_score": 0.27119471729837735,
|
| 102 |
-
"missing_tasks": []
|
| 103 |
-
},
|
| 104 |
-
"Llama_3_2_11B": {
|
| 105 |
-
"num_eval_tasks": 273,
|
| 106 |
-
"num_eval_samples": 4116,
|
| 107 |
-
"num_not_eval_samples": 0,
|
| 108 |
-
"num_total_samples": 4377,
|
| 109 |
-
"macro_mean_score": 0.20789144960796493,
|
| 110 |
-
"micro_mean_score": 0.20163641703273802,
|
| 111 |
-
"missing_tasks": []
|
| 112 |
-
},
|
| 113 |
-
"MiniCPM_v2.6": {
|
| 114 |
-
"num_eval_tasks": 273,
|
| 115 |
-
"num_eval_samples": 4116,
|
| 116 |
-
"num_not_eval_samples": 0,
|
| 117 |
-
"num_total_samples": 4377,
|
| 118 |
-
"macro_mean_score": 0.23230765810722817,
|
| 119 |
-
"micro_mean_score": 0.22684118052665975,
|
| 120 |
-
"missing_tasks": []
|
| 121 |
-
},
|
| 122 |
-
"Molmo_72B": {
|
| 123 |
-
"num_eval_tasks": 270,
|
| 124 |
-
"num_eval_samples": 4073,
|
| 125 |
-
"num_not_eval_samples": 0,
|
| 126 |
-
"num_total_samples": 4331,
|
| 127 |
-
"macro_mean_score": 0.36480000609384927,
|
| 128 |
-
"micro_mean_score": 0.36205779758110807,
|
| 129 |
-
"missing_tasks": [
|
| 130 |
-
"table_understanding",
|
| 131 |
-
"MMSoc_Misinformation_PolitiFact",
|
| 132 |
-
"planning_screenshot_termes"
|
| 133 |
-
]
|
| 134 |
-
},
|
| 135 |
-
"Molmo_7B_D": {
|
| 136 |
-
"num_eval_tasks": 272,
|
| 137 |
-
"num_eval_samples": 4102,
|
| 138 |
-
"num_not_eval_samples": 0,
|
| 139 |
-
"num_total_samples": 4362,
|
| 140 |
-
"macro_mean_score": 0.2098088446992518,
|
| 141 |
-
"micro_mean_score": 0.20550929661464645,
|
| 142 |
-
"missing_tasks": [
|
| 143 |
-
"MMSoc_Misinformation_PolitiFact"
|
| 144 |
-
]
|
| 145 |
-
},
|
| 146 |
-
"NVLM": {
|
| 147 |
-
"num_eval_tasks": 273,
|
| 148 |
-
"num_eval_samples": 4116,
|
| 149 |
-
"num_not_eval_samples": 0,
|
| 150 |
-
"num_total_samples": 4377,
|
| 151 |
-
"macro_mean_score": 0.32989872890926025,
|
| 152 |
-
"micro_mean_score": 0.32315683713111915,
|
| 153 |
-
"missing_tasks": []
|
| 154 |
-
},
|
| 155 |
-
"POINTS_7B": {
|
| 156 |
-
"num_eval_tasks": 273,
|
| 157 |
-
"num_eval_samples": 4116,
|
| 158 |
-
"num_not_eval_samples": 0,
|
| 159 |
-
"num_total_samples": 4377,
|
| 160 |
-
"macro_mean_score": 0.25511317681632334,
|
| 161 |
-
"micro_mean_score": 0.24927711632415062,
|
| 162 |
-
"missing_tasks": []
|
| 163 |
-
},
|
| 164 |
-
"Phi-3.5-vision": {
|
| 165 |
-
"num_eval_tasks": 273,
|
| 166 |
-
"num_eval_samples": 4116,
|
| 167 |
-
"num_not_eval_samples": 0,
|
| 168 |
-
"num_total_samples": 4377,
|
| 169 |
-
"macro_mean_score": 0.2561274958722834,
|
| 170 |
-
"micro_mean_score": 0.2504214576875906,
|
| 171 |
-
"missing_tasks": []
|
| 172 |
-
},
|
| 173 |
-
"Pixtral_12B": {
|
| 174 |
-
"num_eval_tasks": 273,
|
| 175 |
-
"num_eval_samples": 4116,
|
| 176 |
-
"num_not_eval_samples": 0,
|
| 177 |
-
"num_total_samples": 4377,
|
| 178 |
-
"macro_mean_score": 0.3436942439614412,
|
| 179 |
-
"micro_mean_score": 0.3373564384613738,
|
| 180 |
-
"missing_tasks": []
|
| 181 |
-
},
|
| 182 |
-
"Qwen2_VL_2B": {
|
| 183 |
-
"num_eval_tasks": 273,
|
| 184 |
-
"num_eval_samples": 4116,
|
| 185 |
-
"num_not_eval_samples": 0,
|
| 186 |
-
"num_total_samples": 4377,
|
| 187 |
-
"macro_mean_score": 0.22787906973244856,
|
| 188 |
-
"micro_mean_score": 0.2234748515064842,
|
| 189 |
-
"missing_tasks": []
|
| 190 |
-
},
|
| 191 |
-
"Qwen2_VL_72B": {
|
| 192 |
-
"num_eval_tasks": 273,
|
| 193 |
-
"num_eval_samples": 4116,
|
| 194 |
-
"num_not_eval_samples": 0,
|
| 195 |
-
"num_total_samples": 4377,
|
| 196 |
-
"macro_mean_score": 0.4730536307784527,
|
| 197 |
-
"micro_mean_score": 0.4659830915476831,
|
| 198 |
-
"missing_tasks": []
|
| 199 |
-
},
|
| 200 |
-
"Qwen2_VL_7B": {
|
| 201 |
-
"num_eval_tasks": 273,
|
| 202 |
-
"num_eval_samples": 4116,
|
| 203 |
-
"num_not_eval_samples": 0,
|
| 204 |
-
"num_total_samples": 4377,
|
| 205 |
-
"macro_mean_score": 0.3538656561495699,
|
| 206 |
-
"micro_mean_score": 0.34581250459157137,
|
| 207 |
-
"missing_tasks": []
|
| 208 |
-
},
|
| 209 |
-
"llava_onevision_72B": {
|
| 210 |
-
"num_eval_tasks": 273,
|
| 211 |
-
"num_eval_samples": 4116,
|
| 212 |
-
"num_not_eval_samples": 0,
|
| 213 |
-
"num_total_samples": 4377,
|
| 214 |
-
"macro_mean_score": 0.312618242621264,
|
| 215 |
-
"micro_mean_score": 0.3098623876487132,
|
| 216 |
-
"missing_tasks": []
|
| 217 |
-
},
|
| 218 |
-
"llava_onevision_7B": {
|
| 219 |
-
"num_eval_tasks": 273,
|
| 220 |
-
"num_eval_samples": 4116,
|
| 221 |
-
"num_not_eval_samples": 0,
|
| 222 |
-
"num_total_samples": 4377,
|
| 223 |
-
"macro_mean_score": 0.23683339637631812,
|
| 224 |
-
"micro_mean_score": 0.23283041278687175,
|
| 225 |
-
"missing_tasks": []
|
| 226 |
-
}
|
| 227 |
-
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
static/eval_results/Default/all_model_keywords_stats.json
CHANGED
|
@@ -239,25 +239,25 @@
|
|
| 239 |
"count": 303,
|
| 240 |
"num_samples": 4755,
|
| 241 |
"tasks": [],
|
| 242 |
-
"average_score": 0.
|
| 243 |
},
|
| 244 |
"Text Recognition (OCR)": {
|
| 245 |
"count": 137,
|
| 246 |
"num_samples": 2239,
|
| 247 |
"tasks": [],
|
| 248 |
-
"average_score": 0.
|
| 249 |
},
|
| 250 |
"Language Understanding and Generation": {
|
| 251 |
"count": 154,
|
| 252 |
"num_samples": 2509,
|
| 253 |
"tasks": [],
|
| 254 |
-
"average_score": 0.
|
| 255 |
},
|
| 256 |
"Scene and Event Understanding": {
|
| 257 |
"count": 154,
|
| 258 |
"num_samples": 2467,
|
| 259 |
"tasks": [],
|
| 260 |
-
"average_score": 0.
|
| 261 |
},
|
| 262 |
"Mathematical and Logical Reasoning": {
|
| 263 |
"count": 109,
|
|
@@ -269,7 +269,7 @@
|
|
| 269 |
"count": 51,
|
| 270 |
"num_samples": 855,
|
| 271 |
"tasks": [],
|
| 272 |
-
"average_score": 0.
|
| 273 |
},
|
| 274 |
"Ethical and Safety Reasoning": {
|
| 275 |
"count": 15,
|
|
@@ -301,7 +301,7 @@
|
|
| 301 |
"count": 93,
|
| 302 |
"num_samples": 1517,
|
| 303 |
"tasks": [],
|
| 304 |
-
"average_score": 0.
|
| 305 |
},
|
| 306 |
"Text-Based Images and Documents": {
|
| 307 |
"count": 82,
|
|
@@ -331,7 +331,7 @@
|
|
| 331 |
"count": 143,
|
| 332 |
"num_samples": 2248,
|
| 333 |
"tasks": [],
|
| 334 |
-
"average_score": 0.
|
| 335 |
},
|
| 336 |
"3D Models and Aerial Imagery": {
|
| 337 |
"count": 11,
|
|
@@ -351,7 +351,7 @@
|
|
| 351 |
"count": 110,
|
| 352 |
"num_samples": 1714,
|
| 353 |
"tasks": [],
|
| 354 |
-
"average_score": 0.
|
| 355 |
},
|
| 356 |
"exact_text": {
|
| 357 |
"count": 83,
|
|
@@ -389,7 +389,7 @@
|
|
| 389 |
"count": 41,
|
| 390 |
"num_samples": 623,
|
| 391 |
"tasks": [],
|
| 392 |
-
"average_score": 0.
|
| 393 |
},
|
| 394 |
"1-image": {
|
| 395 |
"count": 315,
|
|
@@ -413,7 +413,7 @@
|
|
| 413 |
"count": 51,
|
| 414 |
"num_samples": 802,
|
| 415 |
"tasks": [],
|
| 416 |
-
"average_score": 0.
|
| 417 |
}
|
| 418 |
},
|
| 419 |
"app": {
|
|
@@ -421,7 +421,7 @@
|
|
| 421 |
"count": 72,
|
| 422 |
"num_samples": 1124,
|
| 423 |
"tasks": [],
|
| 424 |
-
"average_score": 0.
|
| 425 |
},
|
| 426 |
"Planning": {
|
| 427 |
"count": 78,
|
|
@@ -457,7 +457,7 @@
|
|
| 457 |
"count": 97,
|
| 458 |
"num_samples": 1605,
|
| 459 |
"tasks": [],
|
| 460 |
-
"average_score": 0.
|
| 461 |
},
|
| 462 |
"Mathematics": {
|
| 463 |
"count": 33,
|
|
@@ -479,13 +479,13 @@
|
|
| 479 |
"count": 137,
|
| 480 |
"num_samples": 2239,
|
| 481 |
"tasks": [],
|
| 482 |
-
"average_score": 0.
|
| 483 |
},
|
| 484 |
"Language Understanding and Generation": {
|
| 485 |
"count": 154,
|
| 486 |
"num_samples": 2509,
|
| 487 |
"tasks": [],
|
| 488 |
-
"average_score": 0.
|
| 489 |
},
|
| 490 |
"Scene and Event Understanding": {
|
| 491 |
"count": 154,
|
|
@@ -535,7 +535,7 @@
|
|
| 535 |
"count": 93,
|
| 536 |
"num_samples": 1517,
|
| 537 |
"tasks": [],
|
| 538 |
-
"average_score": 0.
|
| 539 |
},
|
| 540 |
"Text-Based Images and Documents": {
|
| 541 |
"count": 82,
|
|
@@ -585,7 +585,7 @@
|
|
| 585 |
"count": 110,
|
| 586 |
"num_samples": 1714,
|
| 587 |
"tasks": [],
|
| 588 |
-
"average_score": 0.
|
| 589 |
},
|
| 590 |
"exact_text": {
|
| 591 |
"count": 83,
|
|
@@ -623,7 +623,7 @@
|
|
| 623 |
"count": 41,
|
| 624 |
"num_samples": 623,
|
| 625 |
"tasks": [],
|
| 626 |
-
"average_score": 0.
|
| 627 |
},
|
| 628 |
"1-image": {
|
| 629 |
"count": 315,
|
|
@@ -655,7 +655,7 @@
|
|
| 655 |
"count": 72,
|
| 656 |
"num_samples": 1124,
|
| 657 |
"tasks": [],
|
| 658 |
-
"average_score": 0.
|
| 659 |
},
|
| 660 |
"Planning": {
|
| 661 |
"count": 78,
|
|
@@ -713,13 +713,13 @@
|
|
| 713 |
"count": 137,
|
| 714 |
"num_samples": 2239,
|
| 715 |
"tasks": [],
|
| 716 |
-
"average_score": 0.
|
| 717 |
},
|
| 718 |
"Language Understanding and Generation": {
|
| 719 |
"count": 154,
|
| 720 |
"num_samples": 2509,
|
| 721 |
"tasks": [],
|
| 722 |
-
"average_score": 0.
|
| 723 |
},
|
| 724 |
"Scene and Event Understanding": {
|
| 725 |
"count": 154,
|
|
@@ -769,7 +769,7 @@
|
|
| 769 |
"count": 93,
|
| 770 |
"num_samples": 1517,
|
| 771 |
"tasks": [],
|
| 772 |
-
"average_score": 0.
|
| 773 |
},
|
| 774 |
"Text-Based Images and Documents": {
|
| 775 |
"count": 82,
|
|
@@ -819,7 +819,7 @@
|
|
| 819 |
"count": 110,
|
| 820 |
"num_samples": 1714,
|
| 821 |
"tasks": [],
|
| 822 |
-
"average_score": 0.
|
| 823 |
},
|
| 824 |
"exact_text": {
|
| 825 |
"count": 83,
|
|
@@ -857,7 +857,7 @@
|
|
| 857 |
"count": 41,
|
| 858 |
"num_samples": 623,
|
| 859 |
"tasks": [],
|
| 860 |
-
"average_score": 0.
|
| 861 |
},
|
| 862 |
"1-image": {
|
| 863 |
"count": 315,
|
|
@@ -889,7 +889,7 @@
|
|
| 889 |
"count": 72,
|
| 890 |
"num_samples": 1124,
|
| 891 |
"tasks": [],
|
| 892 |
-
"average_score": 0.
|
| 893 |
},
|
| 894 |
"Planning": {
|
| 895 |
"count": 78,
|
|
@@ -1175,25 +1175,25 @@
|
|
| 1175 |
"count": 303,
|
| 1176 |
"num_samples": 4755,
|
| 1177 |
"tasks": [],
|
| 1178 |
-
"average_score": 0.
|
| 1179 |
},
|
| 1180 |
"Text Recognition (OCR)": {
|
| 1181 |
"count": 137,
|
| 1182 |
"num_samples": 2239,
|
| 1183 |
"tasks": [],
|
| 1184 |
-
"average_score": 0.
|
| 1185 |
},
|
| 1186 |
"Language Understanding and Generation": {
|
| 1187 |
"count": 154,
|
| 1188 |
"num_samples": 2509,
|
| 1189 |
"tasks": [],
|
| 1190 |
-
"average_score": 0.
|
| 1191 |
},
|
| 1192 |
"Scene and Event Understanding": {
|
| 1193 |
"count": 154,
|
| 1194 |
"num_samples": 2467,
|
| 1195 |
"tasks": [],
|
| 1196 |
-
"average_score": 0.
|
| 1197 |
},
|
| 1198 |
"Mathematical and Logical Reasoning": {
|
| 1199 |
"count": 109,
|
|
@@ -1205,7 +1205,7 @@
|
|
| 1205 |
"count": 51,
|
| 1206 |
"num_samples": 855,
|
| 1207 |
"tasks": [],
|
| 1208 |
-
"average_score": 0.
|
| 1209 |
},
|
| 1210 |
"Ethical and Safety Reasoning": {
|
| 1211 |
"count": 15,
|
|
@@ -1237,7 +1237,7 @@
|
|
| 1237 |
"count": 93,
|
| 1238 |
"num_samples": 1517,
|
| 1239 |
"tasks": [],
|
| 1240 |
-
"average_score": 0.
|
| 1241 |
},
|
| 1242 |
"Text-Based Images and Documents": {
|
| 1243 |
"count": 82,
|
|
@@ -1267,7 +1267,7 @@
|
|
| 1267 |
"count": 143,
|
| 1268 |
"num_samples": 2248,
|
| 1269 |
"tasks": [],
|
| 1270 |
-
"average_score": 0.
|
| 1271 |
},
|
| 1272 |
"3D Models and Aerial Imagery": {
|
| 1273 |
"count": 11,
|
|
@@ -1281,13 +1281,13 @@
|
|
| 1281 |
"count": 98,
|
| 1282 |
"num_samples": 1514,
|
| 1283 |
"tasks": [],
|
| 1284 |
-
"average_score": 0.
|
| 1285 |
},
|
| 1286 |
"structured_output": {
|
| 1287 |
"count": 110,
|
| 1288 |
"num_samples": 1714,
|
| 1289 |
"tasks": [],
|
| 1290 |
-
"average_score": 0.
|
| 1291 |
},
|
| 1292 |
"exact_text": {
|
| 1293 |
"count": 83,
|
|
@@ -1325,13 +1325,13 @@
|
|
| 1325 |
"count": 41,
|
| 1326 |
"num_samples": 623,
|
| 1327 |
"tasks": [],
|
| 1328 |
-
"average_score": 0.
|
| 1329 |
},
|
| 1330 |
"1-image": {
|
| 1331 |
"count": 315,
|
| 1332 |
"num_samples": 5228,
|
| 1333 |
"tasks": [],
|
| 1334 |
-
"average_score": 0.
|
| 1335 |
},
|
| 1336 |
"video": {
|
| 1337 |
"count": 43,
|
|
@@ -1349,7 +1349,7 @@
|
|
| 1349 |
"count": 51,
|
| 1350 |
"num_samples": 802,
|
| 1351 |
"tasks": [],
|
| 1352 |
-
"average_score": 0.
|
| 1353 |
}
|
| 1354 |
},
|
| 1355 |
"app": {
|
|
@@ -1357,7 +1357,7 @@
|
|
| 1357 |
"count": 72,
|
| 1358 |
"num_samples": 1124,
|
| 1359 |
"tasks": [],
|
| 1360 |
-
"average_score": 0.
|
| 1361 |
},
|
| 1362 |
"Planning": {
|
| 1363 |
"count": 78,
|
|
@@ -1375,7 +1375,7 @@
|
|
| 1375 |
"count": 145,
|
| 1376 |
"num_samples": 2313,
|
| 1377 |
"tasks": [],
|
| 1378 |
-
"average_score": 0.
|
| 1379 |
},
|
| 1380 |
"Metrics": {
|
| 1381 |
"count": 20,
|
|
@@ -1393,7 +1393,7 @@
|
|
| 1393 |
"count": 97,
|
| 1394 |
"num_samples": 1605,
|
| 1395 |
"tasks": [],
|
| 1396 |
-
"average_score": 0.
|
| 1397 |
},
|
| 1398 |
"Mathematics": {
|
| 1399 |
"count": 33,
|
|
@@ -1409,25 +1409,25 @@
|
|
| 1409 |
"count": 303,
|
| 1410 |
"num_samples": 4755,
|
| 1411 |
"tasks": [],
|
| 1412 |
-
"average_score": 0.
|
| 1413 |
},
|
| 1414 |
"Text Recognition (OCR)": {
|
| 1415 |
"count": 137,
|
| 1416 |
"num_samples": 2239,
|
| 1417 |
"tasks": [],
|
| 1418 |
-
"average_score": 0.
|
| 1419 |
},
|
| 1420 |
"Language Understanding and Generation": {
|
| 1421 |
"count": 154,
|
| 1422 |
"num_samples": 2509,
|
| 1423 |
"tasks": [],
|
| 1424 |
-
"average_score": 0.
|
| 1425 |
},
|
| 1426 |
"Scene and Event Understanding": {
|
| 1427 |
"count": 154,
|
| 1428 |
"num_samples": 2467,
|
| 1429 |
"tasks": [],
|
| 1430 |
-
"average_score": 0.
|
| 1431 |
},
|
| 1432 |
"Mathematical and Logical Reasoning": {
|
| 1433 |
"count": 109,
|
|
@@ -1439,7 +1439,7 @@
|
|
| 1439 |
"count": 51,
|
| 1440 |
"num_samples": 855,
|
| 1441 |
"tasks": [],
|
| 1442 |
-
"average_score": 0.
|
| 1443 |
},
|
| 1444 |
"Ethical and Safety Reasoning": {
|
| 1445 |
"count": 15,
|
|
@@ -1471,7 +1471,7 @@
|
|
| 1471 |
"count": 93,
|
| 1472 |
"num_samples": 1517,
|
| 1473 |
"tasks": [],
|
| 1474 |
-
"average_score": 0.
|
| 1475 |
},
|
| 1476 |
"Text-Based Images and Documents": {
|
| 1477 |
"count": 82,
|
|
@@ -1501,7 +1501,7 @@
|
|
| 1501 |
"count": 143,
|
| 1502 |
"num_samples": 2248,
|
| 1503 |
"tasks": [],
|
| 1504 |
-
"average_score": 0.
|
| 1505 |
},
|
| 1506 |
"3D Models and Aerial Imagery": {
|
| 1507 |
"count": 11,
|
|
@@ -1515,13 +1515,13 @@
|
|
| 1515 |
"count": 98,
|
| 1516 |
"num_samples": 1514,
|
| 1517 |
"tasks": [],
|
| 1518 |
-
"average_score": 0.
|
| 1519 |
},
|
| 1520 |
"structured_output": {
|
| 1521 |
"count": 110,
|
| 1522 |
"num_samples": 1714,
|
| 1523 |
"tasks": [],
|
| 1524 |
-
"average_score": 0.
|
| 1525 |
},
|
| 1526 |
"exact_text": {
|
| 1527 |
"count": 83,
|
|
@@ -1559,13 +1559,13 @@
|
|
| 1559 |
"count": 41,
|
| 1560 |
"num_samples": 623,
|
| 1561 |
"tasks": [],
|
| 1562 |
-
"average_score": 0.
|
| 1563 |
},
|
| 1564 |
"1-image": {
|
| 1565 |
"count": 315,
|
| 1566 |
"num_samples": 5228,
|
| 1567 |
"tasks": [],
|
| 1568 |
-
"average_score": 0.
|
| 1569 |
},
|
| 1570 |
"video": {
|
| 1571 |
"count": 43,
|
|
@@ -1583,7 +1583,7 @@
|
|
| 1583 |
"count": 51,
|
| 1584 |
"num_samples": 802,
|
| 1585 |
"tasks": [],
|
| 1586 |
-
"average_score": 0.
|
| 1587 |
}
|
| 1588 |
},
|
| 1589 |
"app": {
|
|
@@ -1591,7 +1591,7 @@
|
|
| 1591 |
"count": 72,
|
| 1592 |
"num_samples": 1124,
|
| 1593 |
"tasks": [],
|
| 1594 |
-
"average_score": 0.
|
| 1595 |
},
|
| 1596 |
"Planning": {
|
| 1597 |
"count": 78,
|
|
@@ -1609,7 +1609,7 @@
|
|
| 1609 |
"count": 145,
|
| 1610 |
"num_samples": 2313,
|
| 1611 |
"tasks": [],
|
| 1612 |
-
"average_score": 0.
|
| 1613 |
},
|
| 1614 |
"Metrics": {
|
| 1615 |
"count": 20,
|
|
@@ -1627,7 +1627,7 @@
|
|
| 1627 |
"count": 97,
|
| 1628 |
"num_samples": 1605,
|
| 1629 |
"tasks": [],
|
| 1630 |
-
"average_score": 0.
|
| 1631 |
},
|
| 1632 |
"Mathematics": {
|
| 1633 |
"count": 33,
|
|
@@ -1643,25 +1643,25 @@
|
|
| 1643 |
"count": 303,
|
| 1644 |
"num_samples": 4755,
|
| 1645 |
"tasks": [],
|
| 1646 |
-
"average_score": 0.
|
| 1647 |
},
|
| 1648 |
"Text Recognition (OCR)": {
|
| 1649 |
"count": 137,
|
| 1650 |
"num_samples": 2239,
|
| 1651 |
"tasks": [],
|
| 1652 |
-
"average_score": 0.
|
| 1653 |
},
|
| 1654 |
"Language Understanding and Generation": {
|
| 1655 |
"count": 154,
|
| 1656 |
"num_samples": 2511,
|
| 1657 |
"tasks": [],
|
| 1658 |
-
"average_score": 0.
|
| 1659 |
},
|
| 1660 |
"Scene and Event Understanding": {
|
| 1661 |
"count": 154,
|
| 1662 |
"num_samples": 2469,
|
| 1663 |
"tasks": [],
|
| 1664 |
-
"average_score": 0.
|
| 1665 |
},
|
| 1666 |
"Mathematical and Logical Reasoning": {
|
| 1667 |
"count": 109,
|
|
@@ -1673,7 +1673,7 @@
|
|
| 1673 |
"count": 51,
|
| 1674 |
"num_samples": 855,
|
| 1675 |
"tasks": [],
|
| 1676 |
-
"average_score": 0.
|
| 1677 |
},
|
| 1678 |
"Ethical and Safety Reasoning": {
|
| 1679 |
"count": 15,
|
|
@@ -1705,7 +1705,7 @@
|
|
| 1705 |
"count": 93,
|
| 1706 |
"num_samples": 1517,
|
| 1707 |
"tasks": [],
|
| 1708 |
-
"average_score": 0.
|
| 1709 |
},
|
| 1710 |
"Text-Based Images and Documents": {
|
| 1711 |
"count": 82,
|
|
@@ -1735,7 +1735,7 @@
|
|
| 1735 |
"count": 143,
|
| 1736 |
"num_samples": 2248,
|
| 1737 |
"tasks": [],
|
| 1738 |
-
"average_score": 0.
|
| 1739 |
},
|
| 1740 |
"3D Models and Aerial Imagery": {
|
| 1741 |
"count": 11,
|
|
@@ -1749,13 +1749,13 @@
|
|
| 1749 |
"count": 98,
|
| 1750 |
"num_samples": 1514,
|
| 1751 |
"tasks": [],
|
| 1752 |
-
"average_score": 0.
|
| 1753 |
},
|
| 1754 |
"structured_output": {
|
| 1755 |
"count": 110,
|
| 1756 |
"num_samples": 1714,
|
| 1757 |
"tasks": [],
|
| 1758 |
-
"average_score": 0.
|
| 1759 |
},
|
| 1760 |
"exact_text": {
|
| 1761 |
"count": 83,
|
|
@@ -1793,13 +1793,13 @@
|
|
| 1793 |
"count": 41,
|
| 1794 |
"num_samples": 623,
|
| 1795 |
"tasks": [],
|
| 1796 |
-
"average_score": 0.
|
| 1797 |
},
|
| 1798 |
"1-image": {
|
| 1799 |
"count": 315,
|
| 1800 |
"num_samples": 5228,
|
| 1801 |
"tasks": [],
|
| 1802 |
-
"average_score": 0.
|
| 1803 |
},
|
| 1804 |
"video": {
|
| 1805 |
"count": 43,
|
|
@@ -1817,7 +1817,7 @@
|
|
| 1817 |
"count": 51,
|
| 1818 |
"num_samples": 802,
|
| 1819 |
"tasks": [],
|
| 1820 |
-
"average_score": 0.
|
| 1821 |
}
|
| 1822 |
},
|
| 1823 |
"app": {
|
|
@@ -1825,7 +1825,7 @@
|
|
| 1825 |
"count": 72,
|
| 1826 |
"num_samples": 1124,
|
| 1827 |
"tasks": [],
|
| 1828 |
-
"average_score": 0.
|
| 1829 |
},
|
| 1830 |
"Planning": {
|
| 1831 |
"count": 78,
|
|
@@ -1843,7 +1843,7 @@
|
|
| 1843 |
"count": 145,
|
| 1844 |
"num_samples": 2315,
|
| 1845 |
"tasks": [],
|
| 1846 |
-
"average_score": 0.
|
| 1847 |
},
|
| 1848 |
"Metrics": {
|
| 1849 |
"count": 20,
|
|
@@ -1861,7 +1861,7 @@
|
|
| 1861 |
"count": 97,
|
| 1862 |
"num_samples": 1605,
|
| 1863 |
"tasks": [],
|
| 1864 |
-
"average_score": 0.
|
| 1865 |
},
|
| 1866 |
"Mathematics": {
|
| 1867 |
"count": 33,
|
|
@@ -1883,13 +1883,13 @@
|
|
| 1883 |
"count": 137,
|
| 1884 |
"num_samples": 2239,
|
| 1885 |
"tasks": [],
|
| 1886 |
-
"average_score": 0.
|
| 1887 |
},
|
| 1888 |
"Language Understanding and Generation": {
|
| 1889 |
"count": 154,
|
| 1890 |
"num_samples": 2509,
|
| 1891 |
"tasks": [],
|
| 1892 |
-
"average_score": 0.
|
| 1893 |
},
|
| 1894 |
"Scene and Event Understanding": {
|
| 1895 |
"count": 154,
|
|
@@ -1939,7 +1939,7 @@
|
|
| 1939 |
"count": 93,
|
| 1940 |
"num_samples": 1517,
|
| 1941 |
"tasks": [],
|
| 1942 |
-
"average_score": 0.
|
| 1943 |
},
|
| 1944 |
"Text-Based Images and Documents": {
|
| 1945 |
"count": 82,
|
|
@@ -1969,7 +1969,7 @@
|
|
| 1969 |
"count": 143,
|
| 1970 |
"num_samples": 2248,
|
| 1971 |
"tasks": [],
|
| 1972 |
-
"average_score": 0.
|
| 1973 |
},
|
| 1974 |
"3D Models and Aerial Imagery": {
|
| 1975 |
"count": 11,
|
|
@@ -1983,13 +1983,13 @@
|
|
| 1983 |
"count": 98,
|
| 1984 |
"num_samples": 1514,
|
| 1985 |
"tasks": [],
|
| 1986 |
-
"average_score": 0.
|
| 1987 |
},
|
| 1988 |
"structured_output": {
|
| 1989 |
"count": 110,
|
| 1990 |
"num_samples": 1714,
|
| 1991 |
"tasks": [],
|
| 1992 |
-
"average_score": 0.
|
| 1993 |
},
|
| 1994 |
"exact_text": {
|
| 1995 |
"count": 83,
|
|
@@ -2027,13 +2027,13 @@
|
|
| 2027 |
"count": 41,
|
| 2028 |
"num_samples": 623,
|
| 2029 |
"tasks": [],
|
| 2030 |
-
"average_score": 0.
|
| 2031 |
},
|
| 2032 |
"1-image": {
|
| 2033 |
"count": 315,
|
| 2034 |
"num_samples": 5228,
|
| 2035 |
"tasks": [],
|
| 2036 |
-
"average_score": 0.
|
| 2037 |
},
|
| 2038 |
"video": {
|
| 2039 |
"count": 43,
|
|
@@ -2059,7 +2059,7 @@
|
|
| 2059 |
"count": 72,
|
| 2060 |
"num_samples": 1124,
|
| 2061 |
"tasks": [],
|
| 2062 |
-
"average_score": 0.
|
| 2063 |
},
|
| 2064 |
"Planning": {
|
| 2065 |
"count": 78,
|
|
@@ -2077,7 +2077,7 @@
|
|
| 2077 |
"count": 145,
|
| 2078 |
"num_samples": 2313,
|
| 2079 |
"tasks": [],
|
| 2080 |
-
"average_score": 0.
|
| 2081 |
},
|
| 2082 |
"Metrics": {
|
| 2083 |
"count": 20,
|
|
@@ -2117,13 +2117,13 @@
|
|
| 2117 |
"count": 137,
|
| 2118 |
"num_samples": 2239,
|
| 2119 |
"tasks": [],
|
| 2120 |
-
"average_score": 0.
|
| 2121 |
},
|
| 2122 |
"Language Understanding and Generation": {
|
| 2123 |
"count": 154,
|
| 2124 |
"num_samples": 2509,
|
| 2125 |
"tasks": [],
|
| 2126 |
-
"average_score": 0.
|
| 2127 |
},
|
| 2128 |
"Scene and Event Understanding": {
|
| 2129 |
"count": 154,
|
|
@@ -2173,7 +2173,7 @@
|
|
| 2173 |
"count": 93,
|
| 2174 |
"num_samples": 1517,
|
| 2175 |
"tasks": [],
|
| 2176 |
-
"average_score": 0.
|
| 2177 |
},
|
| 2178 |
"Text-Based Images and Documents": {
|
| 2179 |
"count": 82,
|
|
@@ -2203,7 +2203,7 @@
|
|
| 2203 |
"count": 143,
|
| 2204 |
"num_samples": 2248,
|
| 2205 |
"tasks": [],
|
| 2206 |
-
"average_score": 0.
|
| 2207 |
},
|
| 2208 |
"3D Models and Aerial Imagery": {
|
| 2209 |
"count": 11,
|
|
@@ -2217,13 +2217,13 @@
|
|
| 2217 |
"count": 98,
|
| 2218 |
"num_samples": 1514,
|
| 2219 |
"tasks": [],
|
| 2220 |
-
"average_score": 0.
|
| 2221 |
},
|
| 2222 |
"structured_output": {
|
| 2223 |
"count": 110,
|
| 2224 |
"num_samples": 1714,
|
| 2225 |
"tasks": [],
|
| 2226 |
-
"average_score": 0.
|
| 2227 |
},
|
| 2228 |
"exact_text": {
|
| 2229 |
"count": 83,
|
|
@@ -2261,13 +2261,13 @@
|
|
| 2261 |
"count": 41,
|
| 2262 |
"num_samples": 623,
|
| 2263 |
"tasks": [],
|
| 2264 |
-
"average_score": 0.
|
| 2265 |
},
|
| 2266 |
"1-image": {
|
| 2267 |
"count": 315,
|
| 2268 |
"num_samples": 5228,
|
| 2269 |
"tasks": [],
|
| 2270 |
-
"average_score": 0.
|
| 2271 |
},
|
| 2272 |
"video": {
|
| 2273 |
"count": 43,
|
|
@@ -2293,7 +2293,7 @@
|
|
| 2293 |
"count": 72,
|
| 2294 |
"num_samples": 1124,
|
| 2295 |
"tasks": [],
|
| 2296 |
-
"average_score": 0.
|
| 2297 |
},
|
| 2298 |
"Planning": {
|
| 2299 |
"count": 78,
|
|
@@ -2311,7 +2311,7 @@
|
|
| 2311 |
"count": 145,
|
| 2312 |
"num_samples": 2313,
|
| 2313 |
"tasks": [],
|
| 2314 |
-
"average_score": 0.
|
| 2315 |
},
|
| 2316 |
"Metrics": {
|
| 2317 |
"count": 20,
|
|
@@ -2345,25 +2345,25 @@
|
|
| 2345 |
"count": 303,
|
| 2346 |
"num_samples": 4755,
|
| 2347 |
"tasks": [],
|
| 2348 |
-
"average_score": 0.
|
| 2349 |
},
|
| 2350 |
"Text Recognition (OCR)": {
|
| 2351 |
"count": 137,
|
| 2352 |
"num_samples": 2239,
|
| 2353 |
"tasks": [],
|
| 2354 |
-
"average_score": 0.
|
| 2355 |
},
|
| 2356 |
"Language Understanding and Generation": {
|
| 2357 |
"count": 154,
|
| 2358 |
"num_samples": 2509,
|
| 2359 |
"tasks": [],
|
| 2360 |
-
"average_score": 0.
|
| 2361 |
},
|
| 2362 |
"Scene and Event Understanding": {
|
| 2363 |
"count": 154,
|
| 2364 |
"num_samples": 2467,
|
| 2365 |
"tasks": [],
|
| 2366 |
-
"average_score": 0.
|
| 2367 |
},
|
| 2368 |
"Mathematical and Logical Reasoning": {
|
| 2369 |
"count": 109,
|
|
@@ -2375,7 +2375,7 @@
|
|
| 2375 |
"count": 51,
|
| 2376 |
"num_samples": 855,
|
| 2377 |
"tasks": [],
|
| 2378 |
-
"average_score": 0.
|
| 2379 |
},
|
| 2380 |
"Ethical and Safety Reasoning": {
|
| 2381 |
"count": 15,
|
|
@@ -2407,7 +2407,7 @@
|
|
| 2407 |
"count": 93,
|
| 2408 |
"num_samples": 1517,
|
| 2409 |
"tasks": [],
|
| 2410 |
-
"average_score": 0.
|
| 2411 |
},
|
| 2412 |
"Text-Based Images and Documents": {
|
| 2413 |
"count": 82,
|
|
@@ -2437,7 +2437,7 @@
|
|
| 2437 |
"count": 143,
|
| 2438 |
"num_samples": 2248,
|
| 2439 |
"tasks": [],
|
| 2440 |
-
"average_score": 0.
|
| 2441 |
},
|
| 2442 |
"3D Models and Aerial Imagery": {
|
| 2443 |
"count": 11,
|
|
@@ -2451,13 +2451,13 @@
|
|
| 2451 |
"count": 98,
|
| 2452 |
"num_samples": 1514,
|
| 2453 |
"tasks": [],
|
| 2454 |
-
"average_score": 0.
|
| 2455 |
},
|
| 2456 |
"structured_output": {
|
| 2457 |
"count": 110,
|
| 2458 |
"num_samples": 1714,
|
| 2459 |
"tasks": [],
|
| 2460 |
-
"average_score": 0.
|
| 2461 |
},
|
| 2462 |
"exact_text": {
|
| 2463 |
"count": 83,
|
|
@@ -2495,13 +2495,13 @@
|
|
| 2495 |
"count": 41,
|
| 2496 |
"num_samples": 623,
|
| 2497 |
"tasks": [],
|
| 2498 |
-
"average_score": 0.
|
| 2499 |
},
|
| 2500 |
"1-image": {
|
| 2501 |
"count": 315,
|
| 2502 |
"num_samples": 5228,
|
| 2503 |
"tasks": [],
|
| 2504 |
-
"average_score": 0.
|
| 2505 |
},
|
| 2506 |
"video": {
|
| 2507 |
"count": 43,
|
|
@@ -2519,7 +2519,7 @@
|
|
| 2519 |
"count": 51,
|
| 2520 |
"num_samples": 802,
|
| 2521 |
"tasks": [],
|
| 2522 |
-
"average_score": 0.
|
| 2523 |
}
|
| 2524 |
},
|
| 2525 |
"app": {
|
|
@@ -2527,7 +2527,7 @@
|
|
| 2527 |
"count": 72,
|
| 2528 |
"num_samples": 1124,
|
| 2529 |
"tasks": [],
|
| 2530 |
-
"average_score": 0.
|
| 2531 |
},
|
| 2532 |
"Planning": {
|
| 2533 |
"count": 78,
|
|
@@ -2545,7 +2545,7 @@
|
|
| 2545 |
"count": 145,
|
| 2546 |
"num_samples": 2313,
|
| 2547 |
"tasks": [],
|
| 2548 |
-
"average_score": 0.
|
| 2549 |
},
|
| 2550 |
"Metrics": {
|
| 2551 |
"count": 20,
|
|
@@ -2563,7 +2563,7 @@
|
|
| 2563 |
"count": 97,
|
| 2564 |
"num_samples": 1605,
|
| 2565 |
"tasks": [],
|
| 2566 |
-
"average_score": 0.
|
| 2567 |
},
|
| 2568 |
"Mathematics": {
|
| 2569 |
"count": 33,
|
|
@@ -2585,13 +2585,13 @@
|
|
| 2585 |
"count": 137,
|
| 2586 |
"num_samples": 2239,
|
| 2587 |
"tasks": [],
|
| 2588 |
-
"average_score": 0.
|
| 2589 |
},
|
| 2590 |
"Language Understanding and Generation": {
|
| 2591 |
"count": 154,
|
| 2592 |
"num_samples": 2511,
|
| 2593 |
"tasks": [],
|
| 2594 |
-
"average_score": 0.
|
| 2595 |
},
|
| 2596 |
"Scene and Event Understanding": {
|
| 2597 |
"count": 154,
|
|
@@ -2641,7 +2641,7 @@
|
|
| 2641 |
"count": 93,
|
| 2642 |
"num_samples": 1517,
|
| 2643 |
"tasks": [],
|
| 2644 |
-
"average_score": 0.
|
| 2645 |
},
|
| 2646 |
"Text-Based Images and Documents": {
|
| 2647 |
"count": 82,
|
|
@@ -2671,7 +2671,7 @@
|
|
| 2671 |
"count": 143,
|
| 2672 |
"num_samples": 2248,
|
| 2673 |
"tasks": [],
|
| 2674 |
-
"average_score": 0.
|
| 2675 |
},
|
| 2676 |
"3D Models and Aerial Imagery": {
|
| 2677 |
"count": 11,
|
|
@@ -2685,13 +2685,13 @@
|
|
| 2685 |
"count": 98,
|
| 2686 |
"num_samples": 1514,
|
| 2687 |
"tasks": [],
|
| 2688 |
-
"average_score": 0.
|
| 2689 |
},
|
| 2690 |
"structured_output": {
|
| 2691 |
"count": 110,
|
| 2692 |
"num_samples": 1714,
|
| 2693 |
"tasks": [],
|
| 2694 |
-
"average_score": 0.
|
| 2695 |
},
|
| 2696 |
"exact_text": {
|
| 2697 |
"count": 83,
|
|
@@ -2729,13 +2729,13 @@
|
|
| 2729 |
"count": 41,
|
| 2730 |
"num_samples": 623,
|
| 2731 |
"tasks": [],
|
| 2732 |
-
"average_score": 0.
|
| 2733 |
},
|
| 2734 |
"1-image": {
|
| 2735 |
"count": 315,
|
| 2736 |
"num_samples": 5228,
|
| 2737 |
"tasks": [],
|
| 2738 |
-
"average_score": 0.
|
| 2739 |
},
|
| 2740 |
"video": {
|
| 2741 |
"count": 43,
|
|
@@ -2761,7 +2761,7 @@
|
|
| 2761 |
"count": 72,
|
| 2762 |
"num_samples": 1124,
|
| 2763 |
"tasks": [],
|
| 2764 |
-
"average_score": 0.
|
| 2765 |
},
|
| 2766 |
"Planning": {
|
| 2767 |
"count": 78,
|
|
@@ -2779,7 +2779,7 @@
|
|
| 2779 |
"count": 145,
|
| 2780 |
"num_samples": 2315,
|
| 2781 |
"tasks": [],
|
| 2782 |
-
"average_score": 0.
|
| 2783 |
},
|
| 2784 |
"Metrics": {
|
| 2785 |
"count": 20,
|
|
@@ -2813,25 +2813,25 @@
|
|
| 2813 |
"count": 303,
|
| 2814 |
"num_samples": 4755,
|
| 2815 |
"tasks": [],
|
| 2816 |
-
"average_score": 0.
|
| 2817 |
},
|
| 2818 |
"Text Recognition (OCR)": {
|
| 2819 |
"count": 137,
|
| 2820 |
"num_samples": 2239,
|
| 2821 |
"tasks": [],
|
| 2822 |
-
"average_score": 0.
|
| 2823 |
},
|
| 2824 |
"Language Understanding and Generation": {
|
| 2825 |
"count": 154,
|
| 2826 |
"num_samples": 2509,
|
| 2827 |
"tasks": [],
|
| 2828 |
-
"average_score": 0.
|
| 2829 |
},
|
| 2830 |
"Scene and Event Understanding": {
|
| 2831 |
"count": 154,
|
| 2832 |
"num_samples": 2467,
|
| 2833 |
"tasks": [],
|
| 2834 |
-
"average_score": 0.
|
| 2835 |
},
|
| 2836 |
"Mathematical and Logical Reasoning": {
|
| 2837 |
"count": 109,
|
|
@@ -2843,7 +2843,7 @@
|
|
| 2843 |
"count": 51,
|
| 2844 |
"num_samples": 855,
|
| 2845 |
"tasks": [],
|
| 2846 |
-
"average_score": 0.
|
| 2847 |
},
|
| 2848 |
"Ethical and Safety Reasoning": {
|
| 2849 |
"count": 15,
|
|
@@ -2875,7 +2875,7 @@
|
|
| 2875 |
"count": 93,
|
| 2876 |
"num_samples": 1517,
|
| 2877 |
"tasks": [],
|
| 2878 |
-
"average_score": 0.
|
| 2879 |
},
|
| 2880 |
"Text-Based Images and Documents": {
|
| 2881 |
"count": 82,
|
|
@@ -2905,7 +2905,7 @@
|
|
| 2905 |
"count": 143,
|
| 2906 |
"num_samples": 2248,
|
| 2907 |
"tasks": [],
|
| 2908 |
-
"average_score": 0.
|
| 2909 |
},
|
| 2910 |
"3D Models and Aerial Imagery": {
|
| 2911 |
"count": 11,
|
|
@@ -2919,13 +2919,13 @@
|
|
| 2919 |
"count": 98,
|
| 2920 |
"num_samples": 1514,
|
| 2921 |
"tasks": [],
|
| 2922 |
-
"average_score": 0.
|
| 2923 |
},
|
| 2924 |
"structured_output": {
|
| 2925 |
"count": 110,
|
| 2926 |
"num_samples": 1714,
|
| 2927 |
"tasks": [],
|
| 2928 |
-
"average_score": 0.
|
| 2929 |
},
|
| 2930 |
"exact_text": {
|
| 2931 |
"count": 83,
|
|
@@ -2963,13 +2963,13 @@
|
|
| 2963 |
"count": 41,
|
| 2964 |
"num_samples": 623,
|
| 2965 |
"tasks": [],
|
| 2966 |
-
"average_score": 0.
|
| 2967 |
},
|
| 2968 |
"1-image": {
|
| 2969 |
"count": 315,
|
| 2970 |
"num_samples": 5228,
|
| 2971 |
"tasks": [],
|
| 2972 |
-
"average_score": 0.
|
| 2973 |
},
|
| 2974 |
"video": {
|
| 2975 |
"count": 43,
|
|
@@ -2987,7 +2987,7 @@
|
|
| 2987 |
"count": 51,
|
| 2988 |
"num_samples": 802,
|
| 2989 |
"tasks": [],
|
| 2990 |
-
"average_score": 0.
|
| 2991 |
}
|
| 2992 |
},
|
| 2993 |
"app": {
|
|
@@ -2995,7 +2995,7 @@
|
|
| 2995 |
"count": 72,
|
| 2996 |
"num_samples": 1124,
|
| 2997 |
"tasks": [],
|
| 2998 |
-
"average_score": 0.
|
| 2999 |
},
|
| 3000 |
"Planning": {
|
| 3001 |
"count": 78,
|
|
@@ -3013,7 +3013,7 @@
|
|
| 3013 |
"count": 145,
|
| 3014 |
"num_samples": 2313,
|
| 3015 |
"tasks": [],
|
| 3016 |
-
"average_score": 0.
|
| 3017 |
},
|
| 3018 |
"Metrics": {
|
| 3019 |
"count": 20,
|
|
@@ -3031,7 +3031,7 @@
|
|
| 3031 |
"count": 97,
|
| 3032 |
"num_samples": 1605,
|
| 3033 |
"tasks": [],
|
| 3034 |
-
"average_score": 0.
|
| 3035 |
},
|
| 3036 |
"Mathematics": {
|
| 3037 |
"count": 33,
|
|
@@ -3053,13 +3053,13 @@
|
|
| 3053 |
"count": 137,
|
| 3054 |
"num_samples": 2239,
|
| 3055 |
"tasks": [],
|
| 3056 |
-
"average_score": 0.
|
| 3057 |
},
|
| 3058 |
"Language Understanding and Generation": {
|
| 3059 |
"count": 154,
|
| 3060 |
"num_samples": 2509,
|
| 3061 |
"tasks": [],
|
| 3062 |
-
"average_score": 0.
|
| 3063 |
},
|
| 3064 |
"Scene and Event Understanding": {
|
| 3065 |
"count": 154,
|
|
@@ -3109,7 +3109,7 @@
|
|
| 3109 |
"count": 93,
|
| 3110 |
"num_samples": 1517,
|
| 3111 |
"tasks": [],
|
| 3112 |
-
"average_score": 0.
|
| 3113 |
},
|
| 3114 |
"Text-Based Images and Documents": {
|
| 3115 |
"count": 82,
|
|
@@ -3139,7 +3139,7 @@
|
|
| 3139 |
"count": 143,
|
| 3140 |
"num_samples": 2248,
|
| 3141 |
"tasks": [],
|
| 3142 |
-
"average_score": 0.
|
| 3143 |
},
|
| 3144 |
"3D Models and Aerial Imagery": {
|
| 3145 |
"count": 11,
|
|
@@ -3153,13 +3153,13 @@
|
|
| 3153 |
"count": 98,
|
| 3154 |
"num_samples": 1514,
|
| 3155 |
"tasks": [],
|
| 3156 |
-
"average_score": 0.
|
| 3157 |
},
|
| 3158 |
"structured_output": {
|
| 3159 |
"count": 110,
|
| 3160 |
"num_samples": 1714,
|
| 3161 |
"tasks": [],
|
| 3162 |
-
"average_score": 0.
|
| 3163 |
},
|
| 3164 |
"exact_text": {
|
| 3165 |
"count": 83,
|
|
@@ -3197,13 +3197,13 @@
|
|
| 3197 |
"count": 41,
|
| 3198 |
"num_samples": 623,
|
| 3199 |
"tasks": [],
|
| 3200 |
-
"average_score": 0.
|
| 3201 |
},
|
| 3202 |
"1-image": {
|
| 3203 |
"count": 315,
|
| 3204 |
"num_samples": 5228,
|
| 3205 |
"tasks": [],
|
| 3206 |
-
"average_score": 0.
|
| 3207 |
},
|
| 3208 |
"video": {
|
| 3209 |
"count": 43,
|
|
@@ -3229,7 +3229,7 @@
|
|
| 3229 |
"count": 72,
|
| 3230 |
"num_samples": 1124,
|
| 3231 |
"tasks": [],
|
| 3232 |
-
"average_score": 0.
|
| 3233 |
},
|
| 3234 |
"Planning": {
|
| 3235 |
"count": 78,
|
|
@@ -3247,7 +3247,7 @@
|
|
| 3247 |
"count": 145,
|
| 3248 |
"num_samples": 2313,
|
| 3249 |
"tasks": [],
|
| 3250 |
-
"average_score": 0.
|
| 3251 |
},
|
| 3252 |
"Metrics": {
|
| 3253 |
"count": 20,
|
|
@@ -3281,25 +3281,25 @@
|
|
| 3281 |
"count": 303,
|
| 3282 |
"num_samples": 4755,
|
| 3283 |
"tasks": [],
|
| 3284 |
-
"average_score": 0.
|
| 3285 |
},
|
| 3286 |
"Text Recognition (OCR)": {
|
| 3287 |
"count": 137,
|
| 3288 |
"num_samples": 2239,
|
| 3289 |
"tasks": [],
|
| 3290 |
-
"average_score": 0.
|
| 3291 |
},
|
| 3292 |
"Language Understanding and Generation": {
|
| 3293 |
"count": 154,
|
| 3294 |
"num_samples": 2509,
|
| 3295 |
"tasks": [],
|
| 3296 |
-
"average_score": 0.
|
| 3297 |
},
|
| 3298 |
"Scene and Event Understanding": {
|
| 3299 |
"count": 154,
|
| 3300 |
"num_samples": 2467,
|
| 3301 |
"tasks": [],
|
| 3302 |
-
"average_score": 0.
|
| 3303 |
},
|
| 3304 |
"Mathematical and Logical Reasoning": {
|
| 3305 |
"count": 109,
|
|
@@ -3311,7 +3311,7 @@
|
|
| 3311 |
"count": 51,
|
| 3312 |
"num_samples": 855,
|
| 3313 |
"tasks": [],
|
| 3314 |
-
"average_score": 0.
|
| 3315 |
},
|
| 3316 |
"Ethical and Safety Reasoning": {
|
| 3317 |
"count": 15,
|
|
@@ -3343,7 +3343,7 @@
|
|
| 3343 |
"count": 93,
|
| 3344 |
"num_samples": 1517,
|
| 3345 |
"tasks": [],
|
| 3346 |
-
"average_score": 0.
|
| 3347 |
},
|
| 3348 |
"Text-Based Images and Documents": {
|
| 3349 |
"count": 82,
|
|
@@ -3373,7 +3373,7 @@
|
|
| 3373 |
"count": 143,
|
| 3374 |
"num_samples": 2248,
|
| 3375 |
"tasks": [],
|
| 3376 |
-
"average_score": 0.
|
| 3377 |
},
|
| 3378 |
"3D Models and Aerial Imagery": {
|
| 3379 |
"count": 11,
|
|
@@ -3387,13 +3387,13 @@
|
|
| 3387 |
"count": 98,
|
| 3388 |
"num_samples": 1514,
|
| 3389 |
"tasks": [],
|
| 3390 |
-
"average_score": 0.
|
| 3391 |
},
|
| 3392 |
"structured_output": {
|
| 3393 |
"count": 110,
|
| 3394 |
"num_samples": 1714,
|
| 3395 |
"tasks": [],
|
| 3396 |
-
"average_score": 0.
|
| 3397 |
},
|
| 3398 |
"exact_text": {
|
| 3399 |
"count": 83,
|
|
@@ -3431,13 +3431,13 @@
|
|
| 3431 |
"count": 41,
|
| 3432 |
"num_samples": 623,
|
| 3433 |
"tasks": [],
|
| 3434 |
-
"average_score": 0.
|
| 3435 |
},
|
| 3436 |
"1-image": {
|
| 3437 |
"count": 315,
|
| 3438 |
"num_samples": 5228,
|
| 3439 |
"tasks": [],
|
| 3440 |
-
"average_score": 0.
|
| 3441 |
},
|
| 3442 |
"video": {
|
| 3443 |
"count": 43,
|
|
@@ -3455,7 +3455,7 @@
|
|
| 3455 |
"count": 51,
|
| 3456 |
"num_samples": 802,
|
| 3457 |
"tasks": [],
|
| 3458 |
-
"average_score": 0.
|
| 3459 |
}
|
| 3460 |
},
|
| 3461 |
"app": {
|
|
@@ -3463,7 +3463,7 @@
|
|
| 3463 |
"count": 72,
|
| 3464 |
"num_samples": 1124,
|
| 3465 |
"tasks": [],
|
| 3466 |
-
"average_score": 0.
|
| 3467 |
},
|
| 3468 |
"Planning": {
|
| 3469 |
"count": 78,
|
|
@@ -3481,7 +3481,7 @@
|
|
| 3481 |
"count": 145,
|
| 3482 |
"num_samples": 2313,
|
| 3483 |
"tasks": [],
|
| 3484 |
-
"average_score": 0.
|
| 3485 |
},
|
| 3486 |
"Metrics": {
|
| 3487 |
"count": 20,
|
|
@@ -3499,7 +3499,7 @@
|
|
| 3499 |
"count": 97,
|
| 3500 |
"num_samples": 1605,
|
| 3501 |
"tasks": [],
|
| 3502 |
-
"average_score": 0.
|
| 3503 |
},
|
| 3504 |
"Mathematics": {
|
| 3505 |
"count": 33,
|
|
@@ -3521,13 +3521,13 @@
|
|
| 3521 |
"count": 137,
|
| 3522 |
"num_samples": 2239,
|
| 3523 |
"tasks": [],
|
| 3524 |
-
"average_score": 0.
|
| 3525 |
},
|
| 3526 |
"Language Understanding and Generation": {
|
| 3527 |
"count": 154,
|
| 3528 |
"num_samples": 2509,
|
| 3529 |
"tasks": [],
|
| 3530 |
-
"average_score": 0.
|
| 3531 |
},
|
| 3532 |
"Scene and Event Understanding": {
|
| 3533 |
"count": 154,
|
|
@@ -3607,7 +3607,7 @@
|
|
| 3607 |
"count": 143,
|
| 3608 |
"num_samples": 2248,
|
| 3609 |
"tasks": [],
|
| 3610 |
-
"average_score": 0.
|
| 3611 |
},
|
| 3612 |
"3D Models and Aerial Imagery": {
|
| 3613 |
"count": 11,
|
|
@@ -3621,7 +3621,7 @@
|
|
| 3621 |
"count": 98,
|
| 3622 |
"num_samples": 1514,
|
| 3623 |
"tasks": [],
|
| 3624 |
-
"average_score": 0.
|
| 3625 |
},
|
| 3626 |
"structured_output": {
|
| 3627 |
"count": 110,
|
|
@@ -3671,7 +3671,7 @@
|
|
| 3671 |
"count": 315,
|
| 3672 |
"num_samples": 5228,
|
| 3673 |
"tasks": [],
|
| 3674 |
-
"average_score": 0.
|
| 3675 |
},
|
| 3676 |
"video": {
|
| 3677 |
"count": 43,
|
|
@@ -3715,7 +3715,7 @@
|
|
| 3715 |
"count": 145,
|
| 3716 |
"num_samples": 2313,
|
| 3717 |
"tasks": [],
|
| 3718 |
-
"average_score": 0.
|
| 3719 |
},
|
| 3720 |
"Metrics": {
|
| 3721 |
"count": 20,
|
|
|
|
| 239 |
"count": 303,
|
| 240 |
"num_samples": 4755,
|
| 241 |
"tasks": [],
|
| 242 |
+
"average_score": 0.5202055934299538
|
| 243 |
},
|
| 244 |
"Text Recognition (OCR)": {
|
| 245 |
"count": 137,
|
| 246 |
"num_samples": 2239,
|
| 247 |
"tasks": [],
|
| 248 |
+
"average_score": 0.5017043129027509
|
| 249 |
},
|
| 250 |
"Language Understanding and Generation": {
|
| 251 |
"count": 154,
|
| 252 |
"num_samples": 2509,
|
| 253 |
"tasks": [],
|
| 254 |
+
"average_score": 0.5532599716027446
|
| 255 |
},
|
| 256 |
"Scene and Event Understanding": {
|
| 257 |
"count": 154,
|
| 258 |
"num_samples": 2467,
|
| 259 |
"tasks": [],
|
| 260 |
+
"average_score": 0.546753787203128
|
| 261 |
},
|
| 262 |
"Mathematical and Logical Reasoning": {
|
| 263 |
"count": 109,
|
|
|
|
| 269 |
"count": 51,
|
| 270 |
"num_samples": 855,
|
| 271 |
"tasks": [],
|
| 272 |
+
"average_score": 0.5751012914154264
|
| 273 |
},
|
| 274 |
"Ethical and Safety Reasoning": {
|
| 275 |
"count": 15,
|
|
|
|
| 301 |
"count": 93,
|
| 302 |
"num_samples": 1517,
|
| 303 |
"tasks": [],
|
| 304 |
+
"average_score": 0.4625032188638111
|
| 305 |
},
|
| 306 |
"Text-Based Images and Documents": {
|
| 307 |
"count": 82,
|
|
|
|
| 331 |
"count": 143,
|
| 332 |
"num_samples": 2248,
|
| 333 |
"tasks": [],
|
| 334 |
+
"average_score": 0.55005349042813
|
| 335 |
},
|
| 336 |
"3D Models and Aerial Imagery": {
|
| 337 |
"count": 11,
|
|
|
|
| 351 |
"count": 110,
|
| 352 |
"num_samples": 1714,
|
| 353 |
"tasks": [],
|
| 354 |
+
"average_score": 0.44418591808616864
|
| 355 |
},
|
| 356 |
"exact_text": {
|
| 357 |
"count": 83,
|
|
|
|
| 389 |
"count": 41,
|
| 390 |
"num_samples": 623,
|
| 391 |
"tasks": [],
|
| 392 |
+
"average_score": 0.5370278962809547
|
| 393 |
},
|
| 394 |
"1-image": {
|
| 395 |
"count": 315,
|
|
|
|
| 413 |
"count": 51,
|
| 414 |
"num_samples": 802,
|
| 415 |
"tasks": [],
|
| 416 |
+
"average_score": 0.45544217378728585
|
| 417 |
}
|
| 418 |
},
|
| 419 |
"app": {
|
|
|
|
| 421 |
"count": 72,
|
| 422 |
"num_samples": 1124,
|
| 423 |
"tasks": [],
|
| 424 |
+
"average_score": 0.5421439953094952
|
| 425 |
},
|
| 426 |
"Planning": {
|
| 427 |
"count": 78,
|
|
|
|
| 457 |
"count": 97,
|
| 458 |
"num_samples": 1605,
|
| 459 |
"tasks": [],
|
| 460 |
+
"average_score": 0.5722329455291694
|
| 461 |
},
|
| 462 |
"Mathematics": {
|
| 463 |
"count": 33,
|
|
|
|
| 479 |
"count": 137,
|
| 480 |
"num_samples": 2239,
|
| 481 |
"tasks": [],
|
| 482 |
+
"average_score": 0.4337278553354258
|
| 483 |
},
|
| 484 |
"Language Understanding and Generation": {
|
| 485 |
"count": 154,
|
| 486 |
"num_samples": 2509,
|
| 487 |
"tasks": [],
|
| 488 |
+
"average_score": 0.49947464681475356
|
| 489 |
},
|
| 490 |
"Scene and Event Understanding": {
|
| 491 |
"count": 154,
|
|
|
|
| 535 |
"count": 93,
|
| 536 |
"num_samples": 1517,
|
| 537 |
"tasks": [],
|
| 538 |
+
"average_score": 0.3865262916591035
|
| 539 |
},
|
| 540 |
"Text-Based Images and Documents": {
|
| 541 |
"count": 82,
|
|
|
|
| 585 |
"count": 110,
|
| 586 |
"num_samples": 1714,
|
| 587 |
"tasks": [],
|
| 588 |
+
"average_score": 0.39868324168390534
|
| 589 |
},
|
| 590 |
"exact_text": {
|
| 591 |
"count": 83,
|
|
|
|
| 623 |
"count": 41,
|
| 624 |
"num_samples": 623,
|
| 625 |
"tasks": [],
|
| 626 |
+
"average_score": 0.43653808057103954
|
| 627 |
},
|
| 628 |
"1-image": {
|
| 629 |
"count": 315,
|
|
|
|
| 655 |
"count": 72,
|
| 656 |
"num_samples": 1124,
|
| 657 |
"tasks": [],
|
| 658 |
+
"average_score": 0.46645473820179373
|
| 659 |
},
|
| 660 |
"Planning": {
|
| 661 |
"count": 78,
|
|
|
|
| 713 |
"count": 137,
|
| 714 |
"num_samples": 2239,
|
| 715 |
"tasks": [],
|
| 716 |
+
"average_score": 0.6082834220752651
|
| 717 |
},
|
| 718 |
"Language Understanding and Generation": {
|
| 719 |
"count": 154,
|
| 720 |
"num_samples": 2509,
|
| 721 |
"tasks": [],
|
| 722 |
+
"average_score": 0.5745077617490254
|
| 723 |
},
|
| 724 |
"Scene and Event Understanding": {
|
| 725 |
"count": 154,
|
|
|
|
| 769 |
"count": 93,
|
| 770 |
"num_samples": 1517,
|
| 771 |
"tasks": [],
|
| 772 |
+
"average_score": 0.5691641481808987
|
| 773 |
},
|
| 774 |
"Text-Based Images and Documents": {
|
| 775 |
"count": 82,
|
|
|
|
| 819 |
"count": 110,
|
| 820 |
"num_samples": 1714,
|
| 821 |
"tasks": [],
|
| 822 |
+
"average_score": 0.4971460788134188
|
| 823 |
},
|
| 824 |
"exact_text": {
|
| 825 |
"count": 83,
|
|
|
|
| 857 |
"count": 41,
|
| 858 |
"num_samples": 623,
|
| 859 |
"tasks": [],
|
| 860 |
+
"average_score": 0.5414381873407914
|
| 861 |
},
|
| 862 |
"1-image": {
|
| 863 |
"count": 315,
|
|
|
|
| 889 |
"count": 72,
|
| 890 |
"num_samples": 1124,
|
| 891 |
"tasks": [],
|
| 892 |
+
"average_score": 0.6663170946790707
|
| 893 |
},
|
| 894 |
"Planning": {
|
| 895 |
"count": 78,
|
|
|
|
| 1175 |
"count": 303,
|
| 1176 |
"num_samples": 4755,
|
| 1177 |
"tasks": [],
|
| 1178 |
+
"average_score": 0.4492982787524939
|
| 1179 |
},
|
| 1180 |
"Text Recognition (OCR)": {
|
| 1181 |
"count": 137,
|
| 1182 |
"num_samples": 2239,
|
| 1183 |
"tasks": [],
|
| 1184 |
+
"average_score": 0.49026056071002017
|
| 1185 |
},
|
| 1186 |
"Language Understanding and Generation": {
|
| 1187 |
"count": 154,
|
| 1188 |
"num_samples": 2509,
|
| 1189 |
"tasks": [],
|
| 1190 |
+
"average_score": 0.5168957112681365
|
| 1191 |
},
|
| 1192 |
"Scene and Event Understanding": {
|
| 1193 |
"count": 154,
|
| 1194 |
"num_samples": 2467,
|
| 1195 |
"tasks": [],
|
| 1196 |
+
"average_score": 0.46731791428406805
|
| 1197 |
},
|
| 1198 |
"Mathematical and Logical Reasoning": {
|
| 1199 |
"count": 109,
|
|
|
|
| 1205 |
"count": 51,
|
| 1206 |
"num_samples": 855,
|
| 1207 |
"tasks": [],
|
| 1208 |
+
"average_score": 0.5572925295284307
|
| 1209 |
},
|
| 1210 |
"Ethical and Safety Reasoning": {
|
| 1211 |
"count": 15,
|
|
|
|
| 1237 |
"count": 93,
|
| 1238 |
"num_samples": 1517,
|
| 1239 |
"tasks": [],
|
| 1240 |
+
"average_score": 0.47202628409684394
|
| 1241 |
},
|
| 1242 |
"Text-Based Images and Documents": {
|
| 1243 |
"count": 82,
|
|
|
|
| 1267 |
"count": 143,
|
| 1268 |
"num_samples": 2248,
|
| 1269 |
"tasks": [],
|
| 1270 |
+
"average_score": 0.465175334092545
|
| 1271 |
},
|
| 1272 |
"3D Models and Aerial Imagery": {
|
| 1273 |
"count": 11,
|
|
|
|
| 1281 |
"count": 98,
|
| 1282 |
"num_samples": 1514,
|
| 1283 |
"tasks": [],
|
| 1284 |
+
"average_score": 0.41242028190533997
|
| 1285 |
},
|
| 1286 |
"structured_output": {
|
| 1287 |
"count": 110,
|
| 1288 |
"num_samples": 1714,
|
| 1289 |
"tasks": [],
|
| 1290 |
+
"average_score": 0.3906415365938764
|
| 1291 |
},
|
| 1292 |
"exact_text": {
|
| 1293 |
"count": 83,
|
|
|
|
| 1325 |
"count": 41,
|
| 1326 |
"num_samples": 623,
|
| 1327 |
"tasks": [],
|
| 1328 |
+
"average_score": 0.4305788513381019
|
| 1329 |
},
|
| 1330 |
"1-image": {
|
| 1331 |
"count": 315,
|
| 1332 |
"num_samples": 5228,
|
| 1333 |
"tasks": [],
|
| 1334 |
+
"average_score": 0.46343334374251277
|
| 1335 |
},
|
| 1336 |
"video": {
|
| 1337 |
"count": 43,
|
|
|
|
| 1349 |
"count": 51,
|
| 1350 |
"num_samples": 802,
|
| 1351 |
"tasks": [],
|
| 1352 |
+
"average_score": 0.36981497185070983
|
| 1353 |
}
|
| 1354 |
},
|
| 1355 |
"app": {
|
|
|
|
| 1357 |
"count": 72,
|
| 1358 |
"num_samples": 1124,
|
| 1359 |
"tasks": [],
|
| 1360 |
+
"average_score": 0.5666618234843734
|
| 1361 |
},
|
| 1362 |
"Planning": {
|
| 1363 |
"count": 78,
|
|
|
|
| 1375 |
"count": 145,
|
| 1376 |
"num_samples": 2313,
|
| 1377 |
"tasks": [],
|
| 1378 |
+
"average_score": 0.43590838051817093
|
| 1379 |
},
|
| 1380 |
"Metrics": {
|
| 1381 |
"count": 20,
|
|
|
|
| 1393 |
"count": 97,
|
| 1394 |
"num_samples": 1605,
|
| 1395 |
"tasks": [],
|
| 1396 |
+
"average_score": 0.5399167524341886
|
| 1397 |
},
|
| 1398 |
"Mathematics": {
|
| 1399 |
"count": 33,
|
|
|
|
| 1409 |
"count": 303,
|
| 1410 |
"num_samples": 4755,
|
| 1411 |
"tasks": [],
|
| 1412 |
+
"average_score": 0.49787264809826687
|
| 1413 |
},
|
| 1414 |
"Text Recognition (OCR)": {
|
| 1415 |
"count": 137,
|
| 1416 |
"num_samples": 2239,
|
| 1417 |
"tasks": [],
|
| 1418 |
+
"average_score": 0.5439010430283516
|
| 1419 |
},
|
| 1420 |
"Language Understanding and Generation": {
|
| 1421 |
"count": 154,
|
| 1422 |
"num_samples": 2509,
|
| 1423 |
"tasks": [],
|
| 1424 |
+
"average_score": 0.5392244859385411
|
| 1425 |
},
|
| 1426 |
"Scene and Event Understanding": {
|
| 1427 |
"count": 154,
|
| 1428 |
"num_samples": 2467,
|
| 1429 |
"tasks": [],
|
| 1430 |
+
"average_score": 0.509277882172206
|
| 1431 |
},
|
| 1432 |
"Mathematical and Logical Reasoning": {
|
| 1433 |
"count": 109,
|
|
|
|
| 1439 |
"count": 51,
|
| 1440 |
"num_samples": 855,
|
| 1441 |
"tasks": [],
|
| 1442 |
+
"average_score": 0.5676817981386025
|
| 1443 |
},
|
| 1444 |
"Ethical and Safety Reasoning": {
|
| 1445 |
"count": 15,
|
|
|
|
| 1471 |
"count": 93,
|
| 1472 |
"num_samples": 1517,
|
| 1473 |
"tasks": [],
|
| 1474 |
+
"average_score": 0.5402397677488632
|
| 1475 |
},
|
| 1476 |
"Text-Based Images and Documents": {
|
| 1477 |
"count": 82,
|
|
|
|
| 1501 |
"count": 143,
|
| 1502 |
"num_samples": 2248,
|
| 1503 |
"tasks": [],
|
| 1504 |
+
"average_score": 0.49789939867591104
|
| 1505 |
},
|
| 1506 |
"3D Models and Aerial Imagery": {
|
| 1507 |
"count": 11,
|
|
|
|
| 1515 |
"count": 98,
|
| 1516 |
"num_samples": 1514,
|
| 1517 |
"tasks": [],
|
| 1518 |
+
"average_score": 0.44719815365440824
|
| 1519 |
},
|
| 1520 |
"structured_output": {
|
| 1521 |
"count": 110,
|
| 1522 |
"num_samples": 1714,
|
| 1523 |
"tasks": [],
|
| 1524 |
+
"average_score": 0.4500902736468407
|
| 1525 |
},
|
| 1526 |
"exact_text": {
|
| 1527 |
"count": 83,
|
|
|
|
| 1559 |
"count": 41,
|
| 1560 |
"num_samples": 623,
|
| 1561 |
"tasks": [],
|
| 1562 |
+
"average_score": 0.5468722850464449
|
| 1563 |
},
|
| 1564 |
"1-image": {
|
| 1565 |
"count": 315,
|
| 1566 |
"num_samples": 5228,
|
| 1567 |
"tasks": [],
|
| 1568 |
+
"average_score": 0.4918205178721877
|
| 1569 |
},
|
| 1570 |
"video": {
|
| 1571 |
"count": 43,
|
|
|
|
| 1583 |
"count": 51,
|
| 1584 |
"num_samples": 802,
|
| 1585 |
"tasks": [],
|
| 1586 |
+
"average_score": 0.45176098055218655
|
| 1587 |
}
|
| 1588 |
},
|
| 1589 |
"app": {
|
|
|
|
| 1591 |
"count": 72,
|
| 1592 |
"num_samples": 1124,
|
| 1593 |
"tasks": [],
|
| 1594 |
+
"average_score": 0.5807658773593334
|
| 1595 |
},
|
| 1596 |
"Planning": {
|
| 1597 |
"count": 78,
|
|
|
|
| 1609 |
"count": 145,
|
| 1610 |
"num_samples": 2313,
|
| 1611 |
"tasks": [],
|
| 1612 |
+
"average_score": 0.5362106489630868
|
| 1613 |
},
|
| 1614 |
"Metrics": {
|
| 1615 |
"count": 20,
|
|
|
|
| 1627 |
"count": 97,
|
| 1628 |
"num_samples": 1605,
|
| 1629 |
"tasks": [],
|
| 1630 |
+
"average_score": 0.5166939389651373
|
| 1631 |
},
|
| 1632 |
"Mathematics": {
|
| 1633 |
"count": 33,
|
|
|
|
| 1643 |
"count": 303,
|
| 1644 |
"num_samples": 4755,
|
| 1645 |
"tasks": [],
|
| 1646 |
+
"average_score": 0.3708368629321668
|
| 1647 |
},
|
| 1648 |
"Text Recognition (OCR)": {
|
| 1649 |
"count": 137,
|
| 1650 |
"num_samples": 2239,
|
| 1651 |
"tasks": [],
|
| 1652 |
+
"average_score": 0.40213773918065815
|
| 1653 |
},
|
| 1654 |
"Language Understanding and Generation": {
|
| 1655 |
"count": 154,
|
| 1656 |
"num_samples": 2511,
|
| 1657 |
"tasks": [],
|
| 1658 |
+
"average_score": 0.4034335110538307
|
| 1659 |
},
|
| 1660 |
"Scene and Event Understanding": {
|
| 1661 |
"count": 154,
|
| 1662 |
"num_samples": 2469,
|
| 1663 |
"tasks": [],
|
| 1664 |
+
"average_score": 0.4109909230944937
|
| 1665 |
},
|
| 1666 |
"Mathematical and Logical Reasoning": {
|
| 1667 |
"count": 109,
|
|
|
|
| 1673 |
"count": 51,
|
| 1674 |
"num_samples": 855,
|
| 1675 |
"tasks": [],
|
| 1676 |
+
"average_score": 0.49360878418945336
|
| 1677 |
},
|
| 1678 |
"Ethical and Safety Reasoning": {
|
| 1679 |
"count": 15,
|
|
|
|
| 1705 |
"count": 93,
|
| 1706 |
"num_samples": 1517,
|
| 1707 |
"tasks": [],
|
| 1708 |
+
"average_score": 0.3821046882337143
|
| 1709 |
},
|
| 1710 |
"Text-Based Images and Documents": {
|
| 1711 |
"count": 82,
|
|
|
|
| 1735 |
"count": 143,
|
| 1736 |
"num_samples": 2248,
|
| 1737 |
"tasks": [],
|
| 1738 |
+
"average_score": 0.40660144920567376
|
| 1739 |
},
|
| 1740 |
"3D Models and Aerial Imagery": {
|
| 1741 |
"count": 11,
|
|
|
|
| 1749 |
"count": 98,
|
| 1750 |
"num_samples": 1514,
|
| 1751 |
"tasks": [],
|
| 1752 |
+
"average_score": 0.3430730210869785
|
| 1753 |
},
|
| 1754 |
"structured_output": {
|
| 1755 |
"count": 110,
|
| 1756 |
"num_samples": 1714,
|
| 1757 |
"tasks": [],
|
| 1758 |
+
"average_score": 0.3426196933687219
|
| 1759 |
},
|
| 1760 |
"exact_text": {
|
| 1761 |
"count": 83,
|
|
|
|
| 1793 |
"count": 41,
|
| 1794 |
"num_samples": 623,
|
| 1795 |
"tasks": [],
|
| 1796 |
+
"average_score": 0.37453319457428763
|
| 1797 |
},
|
| 1798 |
"1-image": {
|
| 1799 |
"count": 315,
|
| 1800 |
"num_samples": 5228,
|
| 1801 |
"tasks": [],
|
| 1802 |
+
"average_score": 0.37701588079136955
|
| 1803 |
},
|
| 1804 |
"video": {
|
| 1805 |
"count": 43,
|
|
|
|
| 1817 |
"count": 51,
|
| 1818 |
"num_samples": 802,
|
| 1819 |
"tasks": [],
|
| 1820 |
+
"average_score": 0.33008667136891007
|
| 1821 |
}
|
| 1822 |
},
|
| 1823 |
"app": {
|
|
|
|
| 1825 |
"count": 72,
|
| 1826 |
"num_samples": 1124,
|
| 1827 |
"tasks": [],
|
| 1828 |
+
"average_score": 0.42746758545520747
|
| 1829 |
},
|
| 1830 |
"Planning": {
|
| 1831 |
"count": 78,
|
|
|
|
| 1843 |
"count": 145,
|
| 1844 |
"num_samples": 2315,
|
| 1845 |
"tasks": [],
|
| 1846 |
+
"average_score": 0.40048749993497734
|
| 1847 |
},
|
| 1848 |
"Metrics": {
|
| 1849 |
"count": 20,
|
|
|
|
| 1861 |
"count": 97,
|
| 1862 |
"num_samples": 1605,
|
| 1863 |
"tasks": [],
|
| 1864 |
+
"average_score": 0.4276637093173368
|
| 1865 |
},
|
| 1866 |
"Mathematics": {
|
| 1867 |
"count": 33,
|
|
|
|
| 1883 |
"count": 137,
|
| 1884 |
"num_samples": 2239,
|
| 1885 |
"tasks": [],
|
| 1886 |
+
"average_score": 0.2834675874668524
|
| 1887 |
},
|
| 1888 |
"Language Understanding and Generation": {
|
| 1889 |
"count": 154,
|
| 1890 |
"num_samples": 2509,
|
| 1891 |
"tasks": [],
|
| 1892 |
+
"average_score": 0.3674817002808495
|
| 1893 |
},
|
| 1894 |
"Scene and Event Understanding": {
|
| 1895 |
"count": 154,
|
|
|
|
| 1939 |
"count": 93,
|
| 1940 |
"num_samples": 1517,
|
| 1941 |
"tasks": [],
|
| 1942 |
+
"average_score": 0.23380046931752074
|
| 1943 |
},
|
| 1944 |
"Text-Based Images and Documents": {
|
| 1945 |
"count": 82,
|
|
|
|
| 1969 |
"count": 143,
|
| 1970 |
"num_samples": 2248,
|
| 1971 |
"tasks": [],
|
| 1972 |
+
"average_score": 0.4247591719013819
|
| 1973 |
},
|
| 1974 |
"3D Models and Aerial Imagery": {
|
| 1975 |
"count": 11,
|
|
|
|
| 1983 |
"count": 98,
|
| 1984 |
"num_samples": 1514,
|
| 1985 |
"tasks": [],
|
| 1986 |
+
"average_score": 0.2868275930712835
|
| 1987 |
},
|
| 1988 |
"structured_output": {
|
| 1989 |
"count": 110,
|
| 1990 |
"num_samples": 1714,
|
| 1991 |
"tasks": [],
|
| 1992 |
+
"average_score": 0.259450238500612
|
| 1993 |
},
|
| 1994 |
"exact_text": {
|
| 1995 |
"count": 83,
|
|
|
|
| 2027 |
"count": 41,
|
| 2028 |
"num_samples": 623,
|
| 2029 |
"tasks": [],
|
| 2030 |
+
"average_score": 0.28104747671521785
|
| 2031 |
},
|
| 2032 |
"1-image": {
|
| 2033 |
"count": 315,
|
| 2034 |
"num_samples": 5228,
|
| 2035 |
"tasks": [],
|
| 2036 |
+
"average_score": 0.34840850032295206
|
| 2037 |
},
|
| 2038 |
"video": {
|
| 2039 |
"count": 43,
|
|
|
|
| 2059 |
"count": 72,
|
| 2060 |
"num_samples": 1124,
|
| 2061 |
"tasks": [],
|
| 2062 |
+
"average_score": 0.3076421844825067
|
| 2063 |
},
|
| 2064 |
"Planning": {
|
| 2065 |
"count": 78,
|
|
|
|
| 2077 |
"count": 145,
|
| 2078 |
"num_samples": 2313,
|
| 2079 |
"tasks": [],
|
| 2080 |
+
"average_score": 0.38362780453378204
|
| 2081 |
},
|
| 2082 |
"Metrics": {
|
| 2083 |
"count": 20,
|
|
|
|
| 2117 |
"count": 137,
|
| 2118 |
"num_samples": 2239,
|
| 2119 |
"tasks": [],
|
| 2120 |
+
"average_score": 0.19077168655703208
|
| 2121 |
},
|
| 2122 |
"Language Understanding and Generation": {
|
| 2123 |
"count": 154,
|
| 2124 |
"num_samples": 2509,
|
| 2125 |
"tasks": [],
|
| 2126 |
+
"average_score": 0.2555444562659206
|
| 2127 |
},
|
| 2128 |
"Scene and Event Understanding": {
|
| 2129 |
"count": 154,
|
|
|
|
| 2173 |
"count": 93,
|
| 2174 |
"num_samples": 1517,
|
| 2175 |
"tasks": [],
|
| 2176 |
+
"average_score": 0.1466861610319767
|
| 2177 |
},
|
| 2178 |
"Text-Based Images and Documents": {
|
| 2179 |
"count": 82,
|
|
|
|
| 2203 |
"count": 143,
|
| 2204 |
"num_samples": 2248,
|
| 2205 |
"tasks": [],
|
| 2206 |
+
"average_score": 0.3263378734842879
|
| 2207 |
},
|
| 2208 |
"3D Models and Aerial Imagery": {
|
| 2209 |
"count": 11,
|
|
|
|
| 2217 |
"count": 98,
|
| 2218 |
"num_samples": 1514,
|
| 2219 |
"tasks": [],
|
| 2220 |
+
"average_score": 0.20277804188944173
|
| 2221 |
},
|
| 2222 |
"structured_output": {
|
| 2223 |
"count": 110,
|
| 2224 |
"num_samples": 1714,
|
| 2225 |
"tasks": [],
|
| 2226 |
+
"average_score": 0.18291595756285564
|
| 2227 |
},
|
| 2228 |
"exact_text": {
|
| 2229 |
"count": 83,
|
|
|
|
| 2261 |
"count": 41,
|
| 2262 |
"num_samples": 623,
|
| 2263 |
"tasks": [],
|
| 2264 |
+
"average_score": 0.13803800801858385
|
| 2265 |
},
|
| 2266 |
"1-image": {
|
| 2267 |
"count": 315,
|
| 2268 |
"num_samples": 5228,
|
| 2269 |
"tasks": [],
|
| 2270 |
+
"average_score": 0.2548084764084038
|
| 2271 |
},
|
| 2272 |
"video": {
|
| 2273 |
"count": 43,
|
|
|
|
| 2293 |
"count": 72,
|
| 2294 |
"num_samples": 1124,
|
| 2295 |
"tasks": [],
|
| 2296 |
+
"average_score": 0.19283211154717242
|
| 2297 |
},
|
| 2298 |
"Planning": {
|
| 2299 |
"count": 78,
|
|
|
|
| 2311 |
"count": 145,
|
| 2312 |
"num_samples": 2313,
|
| 2313 |
"tasks": [],
|
| 2314 |
+
"average_score": 0.28505205882578405
|
| 2315 |
},
|
| 2316 |
"Metrics": {
|
| 2317 |
"count": 20,
|
|
|
|
| 2345 |
"count": 303,
|
| 2346 |
"num_samples": 4755,
|
| 2347 |
"tasks": [],
|
| 2348 |
+
"average_score": 0.38193012983650343
|
| 2349 |
},
|
| 2350 |
"Text Recognition (OCR)": {
|
| 2351 |
"count": 137,
|
| 2352 |
"num_samples": 2239,
|
| 2353 |
"tasks": [],
|
| 2354 |
+
"average_score": 0.41315219763443384
|
| 2355 |
},
|
| 2356 |
"Language Understanding and Generation": {
|
| 2357 |
"count": 154,
|
| 2358 |
"num_samples": 2509,
|
| 2359 |
"tasks": [],
|
| 2360 |
+
"average_score": 0.43665980552577693
|
| 2361 |
},
|
| 2362 |
"Scene and Event Understanding": {
|
| 2363 |
"count": 154,
|
| 2364 |
"num_samples": 2467,
|
| 2365 |
"tasks": [],
|
| 2366 |
+
"average_score": 0.4265623936500962
|
| 2367 |
},
|
| 2368 |
"Mathematical and Logical Reasoning": {
|
| 2369 |
"count": 109,
|
|
|
|
| 2375 |
"count": 51,
|
| 2376 |
"num_samples": 855,
|
| 2377 |
"tasks": [],
|
| 2378 |
+
"average_score": 0.5257990949897898
|
| 2379 |
},
|
| 2380 |
"Ethical and Safety Reasoning": {
|
| 2381 |
"count": 15,
|
|
|
|
| 2407 |
"count": 93,
|
| 2408 |
"num_samples": 1517,
|
| 2409 |
"tasks": [],
|
| 2410 |
+
"average_score": 0.3634339625985008
|
| 2411 |
},
|
| 2412 |
"Text-Based Images and Documents": {
|
| 2413 |
"count": 82,
|
|
|
|
| 2437 |
"count": 143,
|
| 2438 |
"num_samples": 2248,
|
| 2439 |
"tasks": [],
|
| 2440 |
+
"average_score": 0.42875248733027654
|
| 2441 |
},
|
| 2442 |
"3D Models and Aerial Imagery": {
|
| 2443 |
"count": 11,
|
|
|
|
| 2451 |
"count": 98,
|
| 2452 |
"num_samples": 1514,
|
| 2453 |
"tasks": [],
|
| 2454 |
+
"average_score": 0.3630499545707523
|
| 2455 |
},
|
| 2456 |
"structured_output": {
|
| 2457 |
"count": 110,
|
| 2458 |
"num_samples": 1714,
|
| 2459 |
"tasks": [],
|
| 2460 |
+
"average_score": 0.3476691827105281
|
| 2461 |
},
|
| 2462 |
"exact_text": {
|
| 2463 |
"count": 83,
|
|
|
|
| 2495 |
"count": 41,
|
| 2496 |
"num_samples": 623,
|
| 2497 |
"tasks": [],
|
| 2498 |
+
"average_score": 0.34771123515123364
|
| 2499 |
},
|
| 2500 |
"1-image": {
|
| 2501 |
"count": 315,
|
| 2502 |
"num_samples": 5228,
|
| 2503 |
"tasks": [],
|
| 2504 |
+
"average_score": 0.4145693044465943
|
| 2505 |
},
|
| 2506 |
"video": {
|
| 2507 |
"count": 43,
|
|
|
|
| 2519 |
"count": 51,
|
| 2520 |
"num_samples": 802,
|
| 2521 |
"tasks": [],
|
| 2522 |
+
"average_score": 0.3153417935059416
|
| 2523 |
}
|
| 2524 |
},
|
| 2525 |
"app": {
|
|
|
|
| 2527 |
"count": 72,
|
| 2528 |
"num_samples": 1124,
|
| 2529 |
"tasks": [],
|
| 2530 |
+
"average_score": 0.4306947454508794
|
| 2531 |
},
|
| 2532 |
"Planning": {
|
| 2533 |
"count": 78,
|
|
|
|
| 2545 |
"count": 145,
|
| 2546 |
"num_samples": 2313,
|
| 2547 |
"tasks": [],
|
| 2548 |
+
"average_score": 0.42202934355552685
|
| 2549 |
},
|
| 2550 |
"Metrics": {
|
| 2551 |
"count": 20,
|
|
|
|
| 2563 |
"count": 97,
|
| 2564 |
"num_samples": 1605,
|
| 2565 |
"tasks": [],
|
| 2566 |
+
"average_score": 0.4625649385962016
|
| 2567 |
},
|
| 2568 |
"Mathematics": {
|
| 2569 |
"count": 33,
|
|
|
|
| 2585 |
"count": 137,
|
| 2586 |
"num_samples": 2239,
|
| 2587 |
"tasks": [],
|
| 2588 |
+
"average_score": 0.280559214034858
|
| 2589 |
},
|
| 2590 |
"Language Understanding and Generation": {
|
| 2591 |
"count": 154,
|
| 2592 |
"num_samples": 2511,
|
| 2593 |
"tasks": [],
|
| 2594 |
+
"average_score": 0.32020728060179815
|
| 2595 |
},
|
| 2596 |
"Scene and Event Understanding": {
|
| 2597 |
"count": 154,
|
|
|
|
| 2641 |
"count": 93,
|
| 2642 |
"num_samples": 1517,
|
| 2643 |
"tasks": [],
|
| 2644 |
+
"average_score": 0.22800928556370195
|
| 2645 |
},
|
| 2646 |
"Text-Based Images and Documents": {
|
| 2647 |
"count": 82,
|
|
|
|
| 2671 |
"count": 143,
|
| 2672 |
"num_samples": 2248,
|
| 2673 |
"tasks": [],
|
| 2674 |
+
"average_score": 0.3388056726588417
|
| 2675 |
},
|
| 2676 |
"3D Models and Aerial Imagery": {
|
| 2677 |
"count": 11,
|
|
|
|
| 2685 |
"count": 98,
|
| 2686 |
"num_samples": 1514,
|
| 2687 |
"tasks": [],
|
| 2688 |
+
"average_score": 0.250804626773504
|
| 2689 |
},
|
| 2690 |
"structured_output": {
|
| 2691 |
"count": 110,
|
| 2692 |
"num_samples": 1714,
|
| 2693 |
"tasks": [],
|
| 2694 |
+
"average_score": 0.2522493284864019
|
| 2695 |
},
|
| 2696 |
"exact_text": {
|
| 2697 |
"count": 83,
|
|
|
|
| 2729 |
"count": 41,
|
| 2730 |
"num_samples": 623,
|
| 2731 |
"tasks": [],
|
| 2732 |
+
"average_score": 0.19872104324302098
|
| 2733 |
},
|
| 2734 |
"1-image": {
|
| 2735 |
"count": 315,
|
| 2736 |
"num_samples": 5228,
|
| 2737 |
"tasks": [],
|
| 2738 |
+
"average_score": 0.30088711082969344
|
| 2739 |
},
|
| 2740 |
"video": {
|
| 2741 |
"count": 43,
|
|
|
|
| 2761 |
"count": 72,
|
| 2762 |
"num_samples": 1124,
|
| 2763 |
"tasks": [],
|
| 2764 |
+
"average_score": 0.29129840423784176
|
| 2765 |
},
|
| 2766 |
"Planning": {
|
| 2767 |
"count": 78,
|
|
|
|
| 2779 |
"count": 145,
|
| 2780 |
"num_samples": 2315,
|
| 2781 |
"tasks": [],
|
| 2782 |
+
"average_score": 0.3214666523378005
|
| 2783 |
},
|
| 2784 |
"Metrics": {
|
| 2785 |
"count": 20,
|
|
|
|
| 2813 |
"count": 303,
|
| 2814 |
"num_samples": 4755,
|
| 2815 |
"tasks": [],
|
| 2816 |
+
"average_score": 0.2604967101191775
|
| 2817 |
},
|
| 2818 |
"Text Recognition (OCR)": {
|
| 2819 |
"count": 137,
|
| 2820 |
"num_samples": 2239,
|
| 2821 |
"tasks": [],
|
| 2822 |
+
"average_score": 0.2500331562865158
|
| 2823 |
},
|
| 2824 |
"Language Understanding and Generation": {
|
| 2825 |
"count": 154,
|
| 2826 |
"num_samples": 2509,
|
| 2827 |
"tasks": [],
|
| 2828 |
+
"average_score": 0.3003169369011028
|
| 2829 |
},
|
| 2830 |
"Scene and Event Understanding": {
|
| 2831 |
"count": 154,
|
| 2832 |
"num_samples": 2467,
|
| 2833 |
"tasks": [],
|
| 2834 |
+
"average_score": 0.31808748114668184
|
| 2835 |
},
|
| 2836 |
"Mathematical and Logical Reasoning": {
|
| 2837 |
"count": 109,
|
|
|
|
| 2843 |
"count": 51,
|
| 2844 |
"num_samples": 855,
|
| 2845 |
"tasks": [],
|
| 2846 |
+
"average_score": 0.40732197204308807
|
| 2847 |
},
|
| 2848 |
"Ethical and Safety Reasoning": {
|
| 2849 |
"count": 15,
|
|
|
|
| 2875 |
"count": 93,
|
| 2876 |
"num_samples": 1517,
|
| 2877 |
"tasks": [],
|
| 2878 |
+
"average_score": 0.21195711598986072
|
| 2879 |
},
|
| 2880 |
"Text-Based Images and Documents": {
|
| 2881 |
"count": 82,
|
|
|
|
| 2905 |
"count": 143,
|
| 2906 |
"num_samples": 2248,
|
| 2907 |
"tasks": [],
|
| 2908 |
+
"average_score": 0.3176880312524649
|
| 2909 |
},
|
| 2910 |
"3D Models and Aerial Imagery": {
|
| 2911 |
"count": 11,
|
|
|
|
| 2919 |
"count": 98,
|
| 2920 |
"num_samples": 1514,
|
| 2921 |
"tasks": [],
|
| 2922 |
+
"average_score": 0.23506388020592064
|
| 2923 |
},
|
| 2924 |
"structured_output": {
|
| 2925 |
"count": 110,
|
| 2926 |
"num_samples": 1714,
|
| 2927 |
"tasks": [],
|
| 2928 |
+
"average_score": 0.1781127776443048
|
| 2929 |
},
|
| 2930 |
"exact_text": {
|
| 2931 |
"count": 83,
|
|
|
|
| 2963 |
"count": 41,
|
| 2964 |
"num_samples": 623,
|
| 2965 |
"tasks": [],
|
| 2966 |
+
"average_score": 0.23596215721092323
|
| 2967 |
},
|
| 2968 |
"1-image": {
|
| 2969 |
"count": 315,
|
| 2970 |
"num_samples": 5228,
|
| 2971 |
"tasks": [],
|
| 2972 |
+
"average_score": 0.26319603880798287
|
| 2973 |
},
|
| 2974 |
"video": {
|
| 2975 |
"count": 43,
|
|
|
|
| 2987 |
"count": 51,
|
| 2988 |
"num_samples": 802,
|
| 2989 |
"tasks": [],
|
| 2990 |
+
"average_score": 0.22288558250834017
|
| 2991 |
}
|
| 2992 |
},
|
| 2993 |
"app": {
|
|
|
|
| 2995 |
"count": 72,
|
| 2996 |
"num_samples": 1124,
|
| 2997 |
"tasks": [],
|
| 2998 |
+
"average_score": 0.2666989364424082
|
| 2999 |
},
|
| 3000 |
"Planning": {
|
| 3001 |
"count": 78,
|
|
|
|
| 3013 |
"count": 145,
|
| 3014 |
"num_samples": 2313,
|
| 3015 |
"tasks": [],
|
| 3016 |
+
"average_score": 0.29243044121840894
|
| 3017 |
},
|
| 3018 |
"Metrics": {
|
| 3019 |
"count": 20,
|
|
|
|
| 3031 |
"count": 97,
|
| 3032 |
"num_samples": 1605,
|
| 3033 |
"tasks": [],
|
| 3034 |
+
"average_score": 0.33187729423141027
|
| 3035 |
},
|
| 3036 |
"Mathematics": {
|
| 3037 |
"count": 33,
|
|
|
|
| 3053 |
"count": 137,
|
| 3054 |
"num_samples": 2239,
|
| 3055 |
"tasks": [],
|
| 3056 |
+
"average_score": 0.2483252111012436
|
| 3057 |
},
|
| 3058 |
"Language Understanding and Generation": {
|
| 3059 |
"count": 154,
|
| 3060 |
"num_samples": 2509,
|
| 3061 |
"tasks": [],
|
| 3062 |
+
"average_score": 0.28732942108098564
|
| 3063 |
},
|
| 3064 |
"Scene and Event Understanding": {
|
| 3065 |
"count": 154,
|
|
|
|
| 3109 |
"count": 93,
|
| 3110 |
"num_samples": 1517,
|
| 3111 |
"tasks": [],
|
| 3112 |
+
"average_score": 0.1865974025588298
|
| 3113 |
},
|
| 3114 |
"Text-Based Images and Documents": {
|
| 3115 |
"count": 82,
|
|
|
|
| 3139 |
"count": 143,
|
| 3140 |
"num_samples": 2248,
|
| 3141 |
"tasks": [],
|
| 3142 |
+
"average_score": 0.3413768635559215
|
| 3143 |
},
|
| 3144 |
"3D Models and Aerial Imagery": {
|
| 3145 |
"count": 11,
|
|
|
|
| 3153 |
"count": 98,
|
| 3154 |
"num_samples": 1514,
|
| 3155 |
"tasks": [],
|
| 3156 |
+
"average_score": 0.2177924712685756
|
| 3157 |
},
|
| 3158 |
"structured_output": {
|
| 3159 |
"count": 110,
|
| 3160 |
"num_samples": 1714,
|
| 3161 |
"tasks": [],
|
| 3162 |
+
"average_score": 0.21443984349574025
|
| 3163 |
},
|
| 3164 |
"exact_text": {
|
| 3165 |
"count": 83,
|
|
|
|
| 3197 |
"count": 41,
|
| 3198 |
"num_samples": 623,
|
| 3199 |
"tasks": [],
|
| 3200 |
+
"average_score": 0.14337869666229008
|
| 3201 |
},
|
| 3202 |
"1-image": {
|
| 3203 |
"count": 315,
|
| 3204 |
"num_samples": 5228,
|
| 3205 |
"tasks": [],
|
| 3206 |
+
"average_score": 0.27790147494714373
|
| 3207 |
},
|
| 3208 |
"video": {
|
| 3209 |
"count": 43,
|
|
|
|
| 3229 |
"count": 72,
|
| 3230 |
"num_samples": 1124,
|
| 3231 |
"tasks": [],
|
| 3232 |
+
"average_score": 0.2237087834389946
|
| 3233 |
},
|
| 3234 |
"Planning": {
|
| 3235 |
"count": 78,
|
|
|
|
| 3247 |
"count": 145,
|
| 3248 |
"num_samples": 2313,
|
| 3249 |
"tasks": [],
|
| 3250 |
+
"average_score": 0.316318567258608
|
| 3251 |
},
|
| 3252 |
"Metrics": {
|
| 3253 |
"count": 20,
|
|
|
|
| 3281 |
"count": 303,
|
| 3282 |
"num_samples": 4755,
|
| 3283 |
"tasks": [],
|
| 3284 |
+
"average_score": 0.3460288961410444
|
| 3285 |
},
|
| 3286 |
"Text Recognition (OCR)": {
|
| 3287 |
"count": 137,
|
| 3288 |
"num_samples": 2239,
|
| 3289 |
"tasks": [],
|
| 3290 |
+
"average_score": 0.3777640755922415
|
| 3291 |
},
|
| 3292 |
"Language Understanding and Generation": {
|
| 3293 |
"count": 154,
|
| 3294 |
"num_samples": 2509,
|
| 3295 |
"tasks": [],
|
| 3296 |
+
"average_score": 0.38299418297106824
|
| 3297 |
},
|
| 3298 |
"Scene and Event Understanding": {
|
| 3299 |
"count": 154,
|
| 3300 |
"num_samples": 2467,
|
| 3301 |
"tasks": [],
|
| 3302 |
+
"average_score": 0.3776722463473817
|
| 3303 |
},
|
| 3304 |
"Mathematical and Logical Reasoning": {
|
| 3305 |
"count": 109,
|
|
|
|
| 3311 |
"count": 51,
|
| 3312 |
"num_samples": 855,
|
| 3313 |
"tasks": [],
|
| 3314 |
+
"average_score": 0.419071767659191
|
| 3315 |
},
|
| 3316 |
"Ethical and Safety Reasoning": {
|
| 3317 |
"count": 15,
|
|
|
|
| 3343 |
"count": 93,
|
| 3344 |
"num_samples": 1517,
|
| 3345 |
"tasks": [],
|
| 3346 |
+
"average_score": 0.3070067338940785
|
| 3347 |
},
|
| 3348 |
"Text-Based Images and Documents": {
|
| 3349 |
"count": 82,
|
|
|
|
| 3373 |
"count": 143,
|
| 3374 |
"num_samples": 2248,
|
| 3375 |
"tasks": [],
|
| 3376 |
+
"average_score": 0.37115973962368864
|
| 3377 |
},
|
| 3378 |
"3D Models and Aerial Imagery": {
|
| 3379 |
"count": 11,
|
|
|
|
| 3387 |
"count": 98,
|
| 3388 |
"num_samples": 1514,
|
| 3389 |
"tasks": [],
|
| 3390 |
+
"average_score": 0.3078181788009137
|
| 3391 |
},
|
| 3392 |
"structured_output": {
|
| 3393 |
"count": 110,
|
| 3394 |
"num_samples": 1714,
|
| 3395 |
"tasks": [],
|
| 3396 |
+
"average_score": 0.3188475653127356
|
| 3397 |
},
|
| 3398 |
"exact_text": {
|
| 3399 |
"count": 83,
|
|
|
|
| 3431 |
"count": 41,
|
| 3432 |
"num_samples": 623,
|
| 3433 |
"tasks": [],
|
| 3434 |
+
"average_score": 0.16642294307267227
|
| 3435 |
},
|
| 3436 |
"1-image": {
|
| 3437 |
"count": 315,
|
| 3438 |
"num_samples": 5228,
|
| 3439 |
"tasks": [],
|
| 3440 |
+
"average_score": 0.37108130557306335
|
| 3441 |
},
|
| 3442 |
"video": {
|
| 3443 |
"count": 43,
|
|
|
|
| 3455 |
"count": 51,
|
| 3456 |
"num_samples": 802,
|
| 3457 |
"tasks": [],
|
| 3458 |
+
"average_score": 0.3104621543981899
|
| 3459 |
}
|
| 3460 |
},
|
| 3461 |
"app": {
|
|
|
|
| 3463 |
"count": 72,
|
| 3464 |
"num_samples": 1124,
|
| 3465 |
"tasks": [],
|
| 3466 |
+
"average_score": 0.4300741596942578
|
| 3467 |
},
|
| 3468 |
"Planning": {
|
| 3469 |
"count": 78,
|
|
|
|
| 3481 |
"count": 145,
|
| 3482 |
"num_samples": 2313,
|
| 3483 |
"tasks": [],
|
| 3484 |
+
"average_score": 0.3892097218585385
|
| 3485 |
},
|
| 3486 |
"Metrics": {
|
| 3487 |
"count": 20,
|
|
|
|
| 3499 |
"count": 97,
|
| 3500 |
"num_samples": 1605,
|
| 3501 |
"tasks": [],
|
| 3502 |
+
"average_score": 0.3809515410188075
|
| 3503 |
},
|
| 3504 |
"Mathematics": {
|
| 3505 |
"count": 33,
|
|
|
|
| 3521 |
"count": 137,
|
| 3522 |
"num_samples": 2239,
|
| 3523 |
"tasks": [],
|
| 3524 |
+
"average_score": 0.14328677752263275
|
| 3525 |
},
|
| 3526 |
"Language Understanding and Generation": {
|
| 3527 |
"count": 154,
|
| 3528 |
"num_samples": 2509,
|
| 3529 |
"tasks": [],
|
| 3530 |
+
"average_score": 0.19646404502647707
|
| 3531 |
},
|
| 3532 |
"Scene and Event Understanding": {
|
| 3533 |
"count": 154,
|
|
|
|
| 3607 |
"count": 143,
|
| 3608 |
"num_samples": 2248,
|
| 3609 |
"tasks": [],
|
| 3610 |
+
"average_score": 0.2485354956932213
|
| 3611 |
},
|
| 3612 |
"3D Models and Aerial Imagery": {
|
| 3613 |
"count": 11,
|
|
|
|
| 3621 |
"count": 98,
|
| 3622 |
"num_samples": 1514,
|
| 3623 |
"tasks": [],
|
| 3624 |
+
"average_score": 0.12417283740525839
|
| 3625 |
},
|
| 3626 |
"structured_output": {
|
| 3627 |
"count": 110,
|
|
|
|
| 3671 |
"count": 315,
|
| 3672 |
"num_samples": 5228,
|
| 3673 |
"tasks": [],
|
| 3674 |
+
"average_score": 0.1960107191983825
|
| 3675 |
},
|
| 3676 |
"video": {
|
| 3677 |
"count": 43,
|
|
|
|
| 3715 |
"count": 145,
|
| 3716 |
"num_samples": 2313,
|
| 3717 |
"tasks": [],
|
| 3718 |
+
"average_score": 0.19899465185565898
|
| 3719 |
},
|
| 3720 |
"Metrics": {
|
| 3721 |
"count": 20,
|
static/eval_results/Default/all_summary.json
CHANGED
|
@@ -4,7 +4,6 @@
|
|
| 4 |
"num_eval_tasks": 440,
|
| 5 |
"num_eval_samples": 6539,
|
| 6 |
"num_not_eval_samples": 0,
|
| 7 |
-
"num_total_samples": 6961,
|
| 8 |
"macro_mean_score": 0.5203440930873326,
|
| 9 |
"micro_mean_score": 0.514302640282204
|
| 10 |
},
|
|
@@ -12,14 +11,12 @@
|
|
| 12 |
"num_eval_tasks": 440,
|
| 13 |
"num_eval_samples": 6539,
|
| 14 |
"num_not_eval_samples": 0,
|
| 15 |
-
"num_total_samples": 6961,
|
| 16 |
"macro_mean_score": 0.5265030595065238,
|
| 17 |
"micro_mean_score": 0.5236338521693411
|
| 18 |
},
|
| 19 |
"open": {
|
| 20 |
"num_eval_tasks": 65,
|
| 21 |
"num_eval_samples": 1163,
|
| 22 |
-
"num_total_samples": 2448,
|
| 23 |
"macro_mean_score": 0.6478225794744895,
|
| 24 |
"micro_mean_score": 0.665391229578676
|
| 25 |
},
|
|
@@ -30,85 +27,75 @@
|
|
| 30 |
"num_eval_tasks": 440,
|
| 31 |
"num_eval_samples": 6539,
|
| 32 |
"num_not_eval_samples": 0,
|
| 33 |
-
"
|
| 34 |
-
"
|
| 35 |
-
"micro_mean_score": 0.46403536258864253
|
| 36 |
},
|
| 37 |
"core_cot": {
|
| 38 |
"num_eval_tasks": 440,
|
| 39 |
"num_eval_samples": 6539,
|
| 40 |
"num_not_eval_samples": 0,
|
| 41 |
-
"
|
| 42 |
-
"
|
| 43 |
-
"micro_mean_score": 0.47581906202211677
|
| 44 |
},
|
| 45 |
"open": {
|
| 46 |
"num_eval_tasks": 65,
|
| 47 |
"num_eval_samples": 1163,
|
| 48 |
-
"num_total_samples": 2448,
|
| 49 |
"macro_mean_score": 0.5858190649927173,
|
| 50 |
"micro_mean_score": 0.6104901117798793
|
| 51 |
},
|
| 52 |
-
"overall_score": 0.
|
| 53 |
},
|
| 54 |
"Gemini_1.5_flash_002": {
|
| 55 |
"core_noncot": {
|
| 56 |
"num_eval_tasks": 440,
|
| 57 |
"num_eval_samples": 6539,
|
| 58 |
"num_not_eval_samples": 0,
|
| 59 |
-
"
|
| 60 |
-
"
|
| 61 |
-
"micro_mean_score": 0.41216971462683855
|
| 62 |
},
|
| 63 |
"core_cot": {
|
| 64 |
"num_eval_tasks": 440,
|
| 65 |
"num_eval_samples": 6539,
|
| 66 |
"num_not_eval_samples": 0,
|
| 67 |
-
"
|
| 68 |
-
"
|
| 69 |
-
"micro_mean_score": 0.41216971462683855
|
| 70 |
},
|
| 71 |
"open": {
|
| 72 |
"num_eval_tasks": 65,
|
| 73 |
"num_eval_samples": 1163,
|
| 74 |
-
"num_total_samples": 2168,
|
| 75 |
"macro_mean_score": 0.5691365176285039,
|
| 76 |
"micro_mean_score": 0.5987532244196045
|
| 77 |
},
|
| 78 |
-
"overall_score": 0.
|
| 79 |
},
|
| 80 |
"Claude_3.5": {
|
| 81 |
"core_noncot": {
|
| 82 |
"num_eval_tasks": 440,
|
| 83 |
"num_eval_samples": 6539,
|
| 84 |
"num_not_eval_samples": 0,
|
| 85 |
-
"
|
| 86 |
-
"
|
| 87 |
-
"micro_mean_score": 0.4798092874490549
|
| 88 |
},
|
| 89 |
"core_cot": {
|
| 90 |
"num_eval_tasks": 440,
|
| 91 |
"num_eval_samples": 6539,
|
| 92 |
"num_not_eval_samples": 0,
|
| 93 |
-
"
|
| 94 |
-
"
|
| 95 |
-
"micro_mean_score": 0.4991559743144323
|
| 96 |
},
|
| 97 |
"open": {
|
| 98 |
"num_eval_tasks": 65,
|
| 99 |
"num_eval_samples": 1163,
|
| 100 |
-
"num_total_samples": 2288,
|
| 101 |
"macro_mean_score": 0.6373907158949892,
|
| 102 |
"micro_mean_score": 0.6569647463456579
|
| 103 |
},
|
| 104 |
-
"overall_score": 0.
|
| 105 |
},
|
| 106 |
"Claude_3.5_new": {
|
| 107 |
"core_noncot": {
|
| 108 |
"num_eval_tasks": 440,
|
| 109 |
"num_eval_samples": 6539,
|
| 110 |
"num_not_eval_samples": 0,
|
| 111 |
-
"num_total_samples": 6961,
|
| 112 |
"macro_mean_score": 0.4919657684484185,
|
| 113 |
"micro_mean_score": 0.4874520567007144
|
| 114 |
},
|
|
@@ -116,14 +103,12 @@
|
|
| 116 |
"num_eval_tasks": 440,
|
| 117 |
"num_eval_samples": 6539,
|
| 118 |
"num_not_eval_samples": 0,
|
| 119 |
-
"num_total_samples": 6961,
|
| 120 |
"macro_mean_score": 0.5259191914020757,
|
| 121 |
"micro_mean_score": 0.5230785894131227
|
| 122 |
},
|
| 123 |
"open": {
|
| 124 |
"num_eval_tasks": 65,
|
| 125 |
"num_eval_samples": 1163,
|
| 126 |
-
"num_total_samples": 1224,
|
| 127 |
"macro_mean_score": 0.6563419761104125,
|
| 128 |
"micro_mean_score": 0.6724419604471196
|
| 129 |
},
|
|
@@ -134,267 +119,236 @@
|
|
| 134 |
"num_eval_tasks": 440,
|
| 135 |
"num_eval_samples": 6539,
|
| 136 |
"num_not_eval_samples": 0,
|
| 137 |
-
"
|
| 138 |
-
"
|
| 139 |
-
"micro_mean_score": 0.392578163407945
|
| 140 |
},
|
| 141 |
"core_cot": {
|
| 142 |
"num_eval_tasks": 440,
|
| 143 |
"num_eval_samples": 6539,
|
| 144 |
"num_not_eval_samples": 0,
|
| 145 |
-
"
|
| 146 |
-
"
|
| 147 |
-
"micro_mean_score": 0.40376078514357017
|
| 148 |
},
|
| 149 |
"open": {
|
| 150 |
"num_eval_tasks": 65,
|
| 151 |
"num_eval_samples": 1163,
|
| 152 |
-
"num_total_samples": 1224,
|
| 153 |
"macro_mean_score": 0.586537827213665,
|
| 154 |
"micro_mean_score": 0.6133276010318144
|
| 155 |
},
|
| 156 |
-
"overall_score": 0.
|
| 157 |
},
|
| 158 |
"Qwen2_VL_72B": {
|
| 159 |
"core_noncot": {
|
| 160 |
"num_eval_tasks": 440,
|
| 161 |
"num_eval_samples": 6539,
|
| 162 |
"num_not_eval_samples": 0,
|
| 163 |
-
"
|
| 164 |
-
"
|
| 165 |
-
"micro_mean_score": 0.4568583770401895
|
| 166 |
},
|
| 167 |
"core_cot": {
|
| 168 |
"num_eval_tasks": 440,
|
| 169 |
"num_eval_samples": 6539,
|
| 170 |
"num_not_eval_samples": 0,
|
| 171 |
-
"
|
| 172 |
-
"
|
| 173 |
-
"micro_mean_score": 0.4487693487093462
|
| 174 |
},
|
| 175 |
"open": {
|
| 176 |
"num_eval_tasks": 65,
|
| 177 |
"num_eval_samples": 1163,
|
| 178 |
-
"num_total_samples": 2448,
|
| 179 |
"macro_mean_score": 0.5639771804231668,
|
| 180 |
"micro_mean_score": 0.5835339638865004
|
| 181 |
},
|
| 182 |
-
"overall_score": 0.
|
| 183 |
},
|
| 184 |
"Qwen2_VL_7B": {
|
| 185 |
"core_noncot": {
|
| 186 |
"num_eval_tasks": 440,
|
| 187 |
"num_eval_samples": 6539,
|
| 188 |
"num_not_eval_samples": 0,
|
| 189 |
-
"
|
| 190 |
-
"
|
| 191 |
-
"micro_mean_score": 0.34344091516995323
|
| 192 |
},
|
| 193 |
"core_cot": {
|
| 194 |
"num_eval_tasks": 440,
|
| 195 |
"num_eval_samples": 6539,
|
| 196 |
"num_not_eval_samples": 0,
|
| 197 |
-
"
|
| 198 |
-
"
|
| 199 |
-
"micro_mean_score": 0.32443422147119677
|
| 200 |
},
|
| 201 |
"open": {
|
| 202 |
"num_eval_tasks": 65,
|
| 203 |
"num_eval_samples": 1170,
|
| 204 |
-
"num_total_samples": 2452,
|
| 205 |
"macro_mean_score": 0.43955105763038577,
|
| 206 |
"micro_mean_score": 0.45508547008546996
|
| 207 |
},
|
| 208 |
-
"overall_score": 0.
|
| 209 |
},
|
| 210 |
"llava_onevision_72B": {
|
| 211 |
"core_noncot": {
|
| 212 |
"num_eval_tasks": 440,
|
| 213 |
"num_eval_samples": 6539,
|
| 214 |
"num_not_eval_samples": 0,
|
| 215 |
-
"
|
| 216 |
-
"
|
| 217 |
-
"micro_mean_score": 0.3173848563095166
|
| 218 |
},
|
| 219 |
"core_cot": {
|
| 220 |
"num_eval_tasks": 440,
|
| 221 |
"num_eval_samples": 6539,
|
| 222 |
"num_not_eval_samples": 0,
|
| 223 |
-
"
|
| 224 |
-
"
|
| 225 |
-
"micro_mean_score": 0.2954433666362564
|
| 226 |
},
|
| 227 |
"open": {
|
| 228 |
"num_eval_tasks": 65,
|
| 229 |
"num_eval_samples": 1163,
|
| 230 |
-
"num_total_samples": 1224,
|
| 231 |
"macro_mean_score": 0.4599484231632498,
|
| 232 |
"micro_mean_score": 0.4850386930352536
|
| 233 |
},
|
| 234 |
-
"overall_score": 0.
|
| 235 |
},
|
| 236 |
"llava_onevision_7B": {
|
| 237 |
"core_noncot": {
|
| 238 |
"num_eval_tasks": 440,
|
| 239 |
"num_eval_samples": 6539,
|
| 240 |
"num_not_eval_samples": 0,
|
| 241 |
-
"
|
| 242 |
-
"
|
| 243 |
-
"micro_mean_score": 0.22222171180488767
|
| 244 |
},
|
| 245 |
"core_cot": {
|
| 246 |
"num_eval_tasks": 440,
|
| 247 |
"num_eval_samples": 6539,
|
| 248 |
"num_not_eval_samples": 0,
|
| 249 |
-
"
|
| 250 |
-
"
|
| 251 |
-
"micro_mean_score": 0.210586172002703
|
| 252 |
},
|
| 253 |
"open": {
|
| 254 |
"num_eval_tasks": 65,
|
| 255 |
"num_eval_samples": 1163,
|
| 256 |
-
"num_total_samples": 2448,
|
| 257 |
"macro_mean_score": 0.33979975321921935,
|
| 258 |
"micro_mean_score": 0.36474634565778147
|
| 259 |
},
|
| 260 |
-
"overall_score": 0.
|
| 261 |
},
|
| 262 |
"InternVL2_76B": {
|
| 263 |
"core_noncot": {
|
| 264 |
"num_eval_tasks": 440,
|
| 265 |
"num_eval_samples": 6539,
|
| 266 |
"num_not_eval_samples": 0,
|
| 267 |
-
"
|
| 268 |
-
"
|
| 269 |
-
"micro_mean_score": 0.3452353155814884
|
| 270 |
},
|
| 271 |
"core_cot": {
|
| 272 |
"num_eval_tasks": 440,
|
| 273 |
"num_eval_samples": 6539,
|
| 274 |
"num_not_eval_samples": 0,
|
| 275 |
-
"
|
| 276 |
-
"
|
| 277 |
-
"micro_mean_score": 0.35043335903915124
|
| 278 |
},
|
| 279 |
"open": {
|
| 280 |
"num_eval_tasks": 65,
|
| 281 |
"num_eval_samples": 1163,
|
| 282 |
-
"num_total_samples": 1224,
|
| 283 |
"macro_mean_score": 0.5192997443033639,
|
| 284 |
"micro_mean_score": 0.5421324161650903
|
| 285 |
},
|
| 286 |
-
"overall_score": 0.
|
| 287 |
},
|
| 288 |
"InternVL2_8B": {
|
| 289 |
"core_noncot": {
|
| 290 |
"num_eval_tasks": 440,
|
| 291 |
"num_eval_samples": 6539,
|
| 292 |
"num_not_eval_samples": 0,
|
| 293 |
-
"
|
| 294 |
-
"
|
| 295 |
-
"micro_mean_score": 0.2543416126895087
|
| 296 |
},
|
| 297 |
"core_cot": {
|
| 298 |
"num_eval_tasks": 440,
|
| 299 |
"num_eval_samples": 6539,
|
| 300 |
"num_not_eval_samples": 0,
|
| 301 |
-
"
|
| 302 |
-
"
|
| 303 |
-
"micro_mean_score": 0.23784634936127952
|
| 304 |
},
|
| 305 |
"open": {
|
| 306 |
"num_eval_tasks": 65,
|
| 307 |
"num_eval_samples": 1165,
|
| 308 |
-
"num_total_samples": 2452,
|
| 309 |
"macro_mean_score": 0.3978571701460552,
|
| 310 |
"micro_mean_score": 0.4108583690987125
|
| 311 |
},
|
| 312 |
-
"overall_score": 0.
|
| 313 |
},
|
| 314 |
"MiniCPM_v2.6": {
|
| 315 |
"core_noncot": {
|
| 316 |
"num_eval_tasks": 440,
|
| 317 |
"num_eval_samples": 6539,
|
| 318 |
"num_not_eval_samples": 0,
|
| 319 |
-
"
|
| 320 |
-
"
|
| 321 |
-
"micro_mean_score": 0.22452805919103805
|
| 322 |
},
|
| 323 |
"core_cot": {
|
| 324 |
"num_eval_tasks": 440,
|
| 325 |
"num_eval_samples": 6539,
|
| 326 |
"num_not_eval_samples": 0,
|
| 327 |
-
"
|
| 328 |
-
"
|
| 329 |
-
"micro_mean_score": 0.2250606411323753
|
| 330 |
},
|
| 331 |
"open": {
|
| 332 |
"num_eval_tasks": 65,
|
| 333 |
"num_eval_samples": 1163,
|
| 334 |
-
"num_total_samples": 2448,
|
| 335 |
"macro_mean_score": 0.41728623355613875,
|
| 336 |
"micro_mean_score": 0.43452278589853827
|
| 337 |
},
|
| 338 |
-
"overall_score": 0.
|
| 339 |
},
|
| 340 |
"Phi-3.5-vision": {
|
| 341 |
"core_noncot": {
|
| 342 |
"num_eval_tasks": 440,
|
| 343 |
"num_eval_samples": 6539,
|
| 344 |
"num_not_eval_samples": 0,
|
| 345 |
-
"
|
| 346 |
-
"
|
| 347 |
-
"micro_mean_score": 0.22932978620408923
|
| 348 |
},
|
| 349 |
"core_cot": {
|
| 350 |
"num_eval_tasks": 440,
|
| 351 |
"num_eval_samples": 6539,
|
| 352 |
"num_not_eval_samples": 0,
|
| 353 |
-
"
|
| 354 |
-
"
|
| 355 |
-
"micro_mean_score": 0.2266573336398296
|
| 356 |
},
|
| 357 |
"open": {
|
| 358 |
"num_eval_tasks": 65,
|
| 359 |
"num_eval_samples": 1163,
|
| 360 |
-
"num_total_samples": 2428,
|
| 361 |
"macro_mean_score": 0.3947914647737769,
|
| 362 |
"micro_mean_score": 0.42459157351676696
|
| 363 |
},
|
| 364 |
-
"overall_score": 0.
|
| 365 |
},
|
| 366 |
"Pixtral_12B": {
|
| 367 |
"core_noncot": {
|
| 368 |
"num_eval_tasks": 440,
|
| 369 |
"num_eval_samples": 6539,
|
| 370 |
"num_not_eval_samples": 0,
|
| 371 |
-
"
|
| 372 |
-
"
|
| 373 |
-
"micro_mean_score": 0.3151734861550665
|
| 374 |
},
|
| 375 |
"core_cot": {
|
| 376 |
"num_eval_tasks": 440,
|
| 377 |
"num_eval_samples": 6539,
|
| 378 |
"num_not_eval_samples": 0,
|
| 379 |
-
"
|
| 380 |
-
"
|
| 381 |
-
"micro_mean_score": 0.30971424472967524
|
| 382 |
},
|
| 383 |
"open": {
|
| 384 |
"num_eval_tasks": 65,
|
| 385 |
"num_eval_samples": 1163,
|
| 386 |
-
"num_total_samples": 1224,
|
| 387 |
"macro_mean_score": 0.4566234428542061,
|
| 388 |
"micro_mean_score": 0.4870593293207223
|
| 389 |
},
|
| 390 |
-
"overall_score": 0.
|
| 391 |
},
|
| 392 |
"Llama_3_2_11B": {
|
| 393 |
"core_noncot": {
|
| 394 |
"num_eval_tasks": 440,
|
| 395 |
"num_eval_samples": 6539,
|
| 396 |
"num_not_eval_samples": 0,
|
| 397 |
-
"num_total_samples": 6961,
|
| 398 |
"macro_mean_score": 0.10044261716549671,
|
| 399 |
"micro_mean_score": 0.09980638766828835
|
| 400 |
},
|
|
@@ -402,25 +356,22 @@
|
|
| 402 |
"num_eval_tasks": 440,
|
| 403 |
"num_eval_samples": 6539,
|
| 404 |
"num_not_eval_samples": 0,
|
| 405 |
-
"
|
| 406 |
-
"
|
| 407 |
-
"micro_mean_score": 0.15794038158731832
|
| 408 |
},
|
| 409 |
"open": {
|
| 410 |
"num_eval_tasks": 65,
|
| 411 |
"num_eval_samples": 1163,
|
| 412 |
-
"num_total_samples": 1224,
|
| 413 |
"macro_mean_score": 0.3173342406187366,
|
| 414 |
"micro_mean_score": 0.3487962166809973
|
| 415 |
},
|
| 416 |
-
"overall_score": 0.
|
| 417 |
},
|
| 418 |
"Idefics3": {
|
| 419 |
"core_noncot": {
|
| 420 |
"num_eval_tasks": 440,
|
| 421 |
"num_eval_samples": 6539,
|
| 422 |
"num_not_eval_samples": 0,
|
| 423 |
-
"num_total_samples": 6961,
|
| 424 |
"macro_mean_score": 0.11118980301103833,
|
| 425 |
"micro_mean_score": 0.11201785633274061
|
| 426 |
},
|
|
@@ -428,14 +379,12 @@
|
|
| 428 |
"num_eval_tasks": 440,
|
| 429 |
"num_eval_samples": 6539,
|
| 430 |
"num_not_eval_samples": 0,
|
| 431 |
-
"num_total_samples": 6961,
|
| 432 |
"macro_mean_score": 0.08956972487602757,
|
| 433 |
"micro_mean_score": 0.08982225274252693
|
| 434 |
},
|
| 435 |
"open": {
|
| 436 |
"num_eval_tasks": 65,
|
| 437 |
"num_eval_samples": 1163,
|
| 438 |
-
"num_total_samples": 2448,
|
| 439 |
"macro_mean_score": 0.3210866162255635,
|
| 440 |
"micro_mean_score": 0.35649183147033553
|
| 441 |
},
|
|
@@ -446,7 +395,6 @@
|
|
| 446 |
"num_eval_tasks": 440,
|
| 447 |
"num_eval_samples": 6539,
|
| 448 |
"num_not_eval_samples": 0,
|
| 449 |
-
"num_total_samples": 6961,
|
| 450 |
"macro_mean_score": 0.30485930718699694,
|
| 451 |
"micro_mean_score": 0.3016713629035311
|
| 452 |
},
|
|
@@ -454,14 +402,12 @@
|
|
| 454 |
"num_eval_tasks": 440,
|
| 455 |
"num_eval_samples": 6539,
|
| 456 |
"num_not_eval_samples": 0,
|
| 457 |
-
"num_total_samples": 6961,
|
| 458 |
"macro_mean_score": 0.289073788209904,
|
| 459 |
"micro_mean_score": 0.2859007507765791
|
| 460 |
},
|
| 461 |
"open": {
|
| 462 |
"num_eval_tasks": 65,
|
| 463 |
"num_eval_samples": 1163,
|
| 464 |
-
"num_total_samples": 1224,
|
| 465 |
"macro_mean_score": 0.5103725263180767,
|
| 466 |
"micro_mean_score": 0.5349957007738607
|
| 467 |
},
|
|
@@ -472,7 +418,6 @@
|
|
| 472 |
"num_eval_tasks": 440,
|
| 473 |
"num_eval_samples": 6539,
|
| 474 |
"num_not_eval_samples": 0,
|
| 475 |
-
"num_total_samples": 6961,
|
| 476 |
"macro_mean_score": 0.2420528895703979,
|
| 477 |
"micro_mean_score": 0.23838419989257642
|
| 478 |
},
|
|
@@ -480,14 +425,12 @@
|
|
| 480 |
"num_eval_tasks": 440,
|
| 481 |
"num_eval_samples": 6539,
|
| 482 |
"num_not_eval_samples": 0,
|
| 483 |
-
"num_total_samples": 6961,
|
| 484 |
"macro_mean_score": 0.21589726765847422,
|
| 485 |
"micro_mean_score": 0.21406043849932396
|
| 486 |
},
|
| 487 |
"open": {
|
| 488 |
"num_eval_tasks": 65,
|
| 489 |
"num_eval_samples": 1163,
|
| 490 |
-
"num_total_samples": 1224,
|
| 491 |
"macro_mean_score": 0.3478114310231307,
|
| 492 |
"micro_mean_score": 0.3947549441100602
|
| 493 |
},
|
|
@@ -498,7 +441,6 @@
|
|
| 498 |
"num_eval_tasks": 440,
|
| 499 |
"num_eval_samples": 6539,
|
| 500 |
"num_not_eval_samples": 0,
|
| 501 |
-
"num_total_samples": 6961,
|
| 502 |
"macro_mean_score": 0.09089701489596874,
|
| 503 |
"micro_mean_score": 0.09036328295381871
|
| 504 |
},
|
|
@@ -506,14 +448,12 @@
|
|
| 506 |
"num_eval_tasks": 440,
|
| 507 |
"num_eval_samples": 6539,
|
| 508 |
"num_not_eval_samples": 0,
|
| 509 |
-
"num_total_samples": 6961,
|
| 510 |
"macro_mean_score": 0.13141974398938763,
|
| 511 |
"micro_mean_score": 0.13063500716262516
|
| 512 |
},
|
| 513 |
"open": {
|
| 514 |
"num_eval_tasks": 65,
|
| 515 |
"num_eval_samples": 1163,
|
| 516 |
-
"num_total_samples": 1224,
|
| 517 |
"macro_mean_score": 0.23864417043743646,
|
| 518 |
"micro_mean_score": 0.24901117798796224
|
| 519 |
},
|
|
@@ -524,7 +464,6 @@
|
|
| 524 |
"num_eval_tasks": 440,
|
| 525 |
"num_eval_samples": 6539,
|
| 526 |
"num_not_eval_samples": 0,
|
| 527 |
-
"num_total_samples": 6961,
|
| 528 |
"macro_mean_score": 0.16448220309703876,
|
| 529 |
"micro_mean_score": 0.1610710186451323
|
| 530 |
},
|
|
@@ -532,14 +471,12 @@
|
|
| 532 |
"num_eval_tasks": 440,
|
| 533 |
"num_eval_samples": 6539,
|
| 534 |
"num_not_eval_samples": 0,
|
| 535 |
-
"num_total_samples": 6961,
|
| 536 |
"macro_mean_score": 0.20877163406364055,
|
| 537 |
"micro_mean_score": 0.20561526268932287
|
| 538 |
},
|
| 539 |
"open": {
|
| 540 |
"num_eval_tasks": 65,
|
| 541 |
"num_eval_samples": 1163,
|
| 542 |
-
"num_total_samples": 1224,
|
| 543 |
"macro_mean_score": 0.3154302566225611,
|
| 544 |
"micro_mean_score": 0.33856405846947557
|
| 545 |
},
|
|
@@ -550,7 +487,6 @@
|
|
| 550 |
"num_eval_tasks": 440,
|
| 551 |
"num_eval_samples": 6539,
|
| 552 |
"num_not_eval_samples": 0,
|
| 553 |
-
"num_total_samples": 6961,
|
| 554 |
"macro_mean_score": 0.16317824309838627,
|
| 555 |
"micro_mean_score": 0.16198837245148487
|
| 556 |
},
|
|
@@ -558,14 +494,12 @@
|
|
| 558 |
"num_eval_tasks": 440,
|
| 559 |
"num_eval_samples": 6539,
|
| 560 |
"num_not_eval_samples": 0,
|
| 561 |
-
"num_total_samples": 6961,
|
| 562 |
"macro_mean_score": 0.159970161379836,
|
| 563 |
"micro_mean_score": 0.15844711671722148
|
| 564 |
},
|
| 565 |
"open": {
|
| 566 |
"num_eval_tasks": 65,
|
| 567 |
"num_eval_samples": 1163,
|
| 568 |
-
"num_total_samples": 1224,
|
| 569 |
"macro_mean_score": 0.24567572098570653,
|
| 570 |
"micro_mean_score": 0.2704213241616509
|
| 571 |
},
|
|
|
|
| 4 |
"num_eval_tasks": 440,
|
| 5 |
"num_eval_samples": 6539,
|
| 6 |
"num_not_eval_samples": 0,
|
|
|
|
| 7 |
"macro_mean_score": 0.5203440930873326,
|
| 8 |
"micro_mean_score": 0.514302640282204
|
| 9 |
},
|
|
|
|
| 11 |
"num_eval_tasks": 440,
|
| 12 |
"num_eval_samples": 6539,
|
| 13 |
"num_not_eval_samples": 0,
|
|
|
|
| 14 |
"macro_mean_score": 0.5265030595065238,
|
| 15 |
"micro_mean_score": 0.5236338521693411
|
| 16 |
},
|
| 17 |
"open": {
|
| 18 |
"num_eval_tasks": 65,
|
| 19 |
"num_eval_samples": 1163,
|
|
|
|
| 20 |
"macro_mean_score": 0.6478225794744895,
|
| 21 |
"micro_mean_score": 0.665391229578676
|
| 22 |
},
|
|
|
|
| 27 |
"num_eval_tasks": 440,
|
| 28 |
"num_eval_samples": 6539,
|
| 29 |
"num_not_eval_samples": 0,
|
| 30 |
+
"macro_mean_score": 0.4699992918320008,
|
| 31 |
+
"micro_mean_score": 0.4651116133689296
|
|
|
|
| 32 |
},
|
| 33 |
"core_cot": {
|
| 34 |
"num_eval_tasks": 440,
|
| 35 |
"num_eval_samples": 6539,
|
| 36 |
"num_not_eval_samples": 0,
|
| 37 |
+
"macro_mean_score": 0.4822473962867704,
|
| 38 |
+
"micro_mean_score": 0.4764805563057179
|
|
|
|
| 39 |
},
|
| 40 |
"open": {
|
| 41 |
"num_eval_tasks": 65,
|
| 42 |
"num_eval_samples": 1163,
|
|
|
|
| 43 |
"macro_mean_score": 0.5858190649927173,
|
| 44 |
"micro_mean_score": 0.6104901117798793
|
| 45 |
},
|
| 46 |
+
"overall_score": 0.4955784031499121
|
| 47 |
},
|
| 48 |
"Gemini_1.5_flash_002": {
|
| 49 |
"core_noncot": {
|
| 50 |
"num_eval_tasks": 440,
|
| 51 |
"num_eval_samples": 6539,
|
| 52 |
"num_not_eval_samples": 0,
|
| 53 |
+
"macro_mean_score": 0.41898948981774853,
|
| 54 |
+
"micro_mean_score": 0.4127376993779598
|
|
|
|
| 55 |
},
|
| 56 |
"core_cot": {
|
| 57 |
"num_eval_tasks": 440,
|
| 58 |
"num_eval_samples": 6539,
|
| 59 |
"num_not_eval_samples": 0,
|
| 60 |
+
"macro_mean_score": 0.4189319021967416,
|
| 61 |
+
"micro_mean_score": 0.41567515414375245
|
|
|
|
| 62 |
},
|
| 63 |
"open": {
|
| 64 |
"num_eval_tasks": 65,
|
| 65 |
"num_eval_samples": 1163,
|
|
|
|
| 66 |
"macro_mean_score": 0.5691365176285039,
|
| 67 |
"micro_mean_score": 0.5987532244196045
|
| 68 |
},
|
| 69 |
+
"overall_score": 0.43831534488249924
|
| 70 |
},
|
| 71 |
"Claude_3.5": {
|
| 72 |
"core_noncot": {
|
| 73 |
"num_eval_tasks": 440,
|
| 74 |
"num_eval_samples": 6539,
|
| 75 |
"num_not_eval_samples": 0,
|
| 76 |
+
"macro_mean_score": 0.48800427486796155,
|
| 77 |
+
"micro_mean_score": 0.4814327812005499
|
|
|
|
| 78 |
},
|
| 79 |
"core_cot": {
|
| 80 |
"num_eval_tasks": 440,
|
| 81 |
"num_eval_samples": 6539,
|
| 82 |
"num_not_eval_samples": 0,
|
| 83 |
+
"macro_mean_score": 0.5040975742801586,
|
| 84 |
+
"micro_mean_score": 0.5002259116666758
|
|
|
|
| 85 |
},
|
| 86 |
"open": {
|
| 87 |
"num_eval_tasks": 65,
|
| 88 |
"num_eval_samples": 1163,
|
|
|
|
| 89 |
"macro_mean_score": 0.6373907158949892,
|
| 90 |
"micro_mean_score": 0.6569647463456579
|
| 91 |
},
|
| 92 |
+
"overall_score": 0.5212541172602853
|
| 93 |
},
|
| 94 |
"Claude_3.5_new": {
|
| 95 |
"core_noncot": {
|
| 96 |
"num_eval_tasks": 440,
|
| 97 |
"num_eval_samples": 6539,
|
| 98 |
"num_not_eval_samples": 0,
|
|
|
|
| 99 |
"macro_mean_score": 0.4919657684484185,
|
| 100 |
"micro_mean_score": 0.4874520567007144
|
| 101 |
},
|
|
|
|
| 103 |
"num_eval_tasks": 440,
|
| 104 |
"num_eval_samples": 6539,
|
| 105 |
"num_not_eval_samples": 0,
|
|
|
|
| 106 |
"macro_mean_score": 0.5259191914020757,
|
| 107 |
"micro_mean_score": 0.5230785894131227
|
| 108 |
},
|
| 109 |
"open": {
|
| 110 |
"num_eval_tasks": 65,
|
| 111 |
"num_eval_samples": 1163,
|
|
|
|
| 112 |
"macro_mean_score": 0.6563419761104125,
|
| 113 |
"micro_mean_score": 0.6724419604471196
|
| 114 |
},
|
|
|
|
| 119 |
"num_eval_tasks": 440,
|
| 120 |
"num_eval_samples": 6539,
|
| 121 |
"num_not_eval_samples": 0,
|
| 122 |
+
"macro_mean_score": 0.39854757130003565,
|
| 123 |
+
"micro_mean_score": 0.3936551517403452
|
|
|
|
| 124 |
},
|
| 125 |
"core_cot": {
|
| 126 |
"num_eval_tasks": 440,
|
| 127 |
"num_eval_samples": 6539,
|
| 128 |
"num_not_eval_samples": 0,
|
| 129 |
+
"macro_mean_score": 0.40767494558789397,
|
| 130 |
+
"micro_mean_score": 0.40431644154143376
|
|
|
|
| 131 |
},
|
| 132 |
"open": {
|
| 133 |
"num_eval_tasks": 65,
|
| 134 |
"num_eval_samples": 1163,
|
|
|
|
| 135 |
"macro_mean_score": 0.586537827213665,
|
| 136 |
"micro_mean_score": 0.6133276010318144
|
| 137 |
},
|
| 138 |
+
"overall_score": 0.43069690064863675
|
| 139 |
},
|
| 140 |
"Qwen2_VL_72B": {
|
| 141 |
"core_noncot": {
|
| 142 |
"num_eval_tasks": 440,
|
| 143 |
"num_eval_samples": 6539,
|
| 144 |
"num_not_eval_samples": 0,
|
| 145 |
+
"macro_mean_score": 0.46406654108789214,
|
| 146 |
+
"micro_mean_score": 0.4584702152011697
|
|
|
|
| 147 |
},
|
| 148 |
"core_cot": {
|
| 149 |
"num_eval_tasks": 440,
|
| 150 |
"num_eval_samples": 6539,
|
| 151 |
"num_not_eval_samples": 0,
|
| 152 |
+
"macro_mean_score": 0.4542376574527161,
|
| 153 |
+
"micro_mean_score": 0.4501201906164793
|
|
|
|
| 154 |
},
|
| 155 |
"open": {
|
| 156 |
"num_eval_tasks": 65,
|
| 157 |
"num_eval_samples": 1163,
|
|
|
|
| 158 |
"macro_mean_score": 0.5639771804231668,
|
| 159 |
"micro_mean_score": 0.5835339638865004
|
| 160 |
},
|
| 161 |
+
"overall_score": 0.4769263263488681
|
| 162 |
},
|
| 163 |
"Qwen2_VL_7B": {
|
| 164 |
"core_noncot": {
|
| 165 |
"num_eval_tasks": 440,
|
| 166 |
"num_eval_samples": 6539,
|
| 167 |
"num_not_eval_samples": 0,
|
| 168 |
+
"macro_mean_score": 0.3480020832611913,
|
| 169 |
+
"micro_mean_score": 0.3441858958345098
|
|
|
|
| 170 |
},
|
| 171 |
"core_cot": {
|
| 172 |
"num_eval_tasks": 440,
|
| 173 |
"num_eval_samples": 6539,
|
| 174 |
"num_not_eval_samples": 0,
|
| 175 |
+
"macro_mean_score": 0.3293449599230247,
|
| 176 |
+
"micro_mean_score": 0.325331493515679
|
|
|
|
| 177 |
},
|
| 178 |
"open": {
|
| 179 |
"num_eval_tasks": 65,
|
| 180 |
"num_eval_samples": 1170,
|
|
|
|
| 181 |
"macro_mean_score": 0.43955105763038577,
|
| 182 |
"micro_mean_score": 0.45508547008546996
|
| 183 |
},
|
| 184 |
+
"overall_score": 0.3597856146156421
|
| 185 |
},
|
| 186 |
"llava_onevision_72B": {
|
| 187 |
"core_noncot": {
|
| 188 |
"num_eval_tasks": 440,
|
| 189 |
"num_eval_samples": 6539,
|
| 190 |
"num_not_eval_samples": 0,
|
| 191 |
+
"macro_mean_score": 0.3199332158220174,
|
| 192 |
+
"micro_mean_score": 0.31770770553892647
|
|
|
|
| 193 |
},
|
| 194 |
"core_cot": {
|
| 195 |
"num_eval_tasks": 440,
|
| 196 |
"num_eval_samples": 6539,
|
| 197 |
"num_not_eval_samples": 0,
|
| 198 |
+
"macro_mean_score": 0.2974368415462532,
|
| 199 |
+
"micro_mean_score": 0.2956217833156672
|
|
|
|
| 200 |
},
|
| 201 |
"open": {
|
| 202 |
"num_eval_tasks": 65,
|
| 203 |
"num_eval_samples": 1163,
|
|
|
|
| 204 |
"macro_mean_score": 0.4599484231632498,
|
| 205 |
"micro_mean_score": 0.4850386930352536
|
| 206 |
},
|
| 207 |
+
"overall_score": 0.33795497518277007
|
| 208 |
},
|
| 209 |
"llava_onevision_7B": {
|
| 210 |
"core_noncot": {
|
| 211 |
"num_eval_tasks": 440,
|
| 212 |
"num_eval_samples": 6539,
|
| 213 |
"num_not_eval_samples": 0,
|
| 214 |
+
"macro_mean_score": 0.22409531510496777,
|
| 215 |
+
"micro_mean_score": 0.22238854298563537
|
|
|
|
| 216 |
},
|
| 217 |
"core_cot": {
|
| 218 |
"num_eval_tasks": 440,
|
| 219 |
"num_eval_samples": 6539,
|
| 220 |
"num_not_eval_samples": 0,
|
| 221 |
+
"macro_mean_score": 0.21362697219149712,
|
| 222 |
+
"micro_mean_score": 0.21073910058505504
|
|
|
|
| 223 |
},
|
| 224 |
"open": {
|
| 225 |
"num_eval_tasks": 65,
|
| 226 |
"num_eval_samples": 1163,
|
|
|
|
| 227 |
"macro_mean_score": 0.33979975321921935,
|
| 228 |
"micro_mean_score": 0.36474634565778147
|
| 229 |
},
|
| 230 |
+
"overall_score": 0.23898796555531696
|
| 231 |
},
|
| 232 |
"InternVL2_76B": {
|
| 233 |
"core_noncot": {
|
| 234 |
"num_eval_tasks": 440,
|
| 235 |
"num_eval_samples": 6539,
|
| 236 |
"num_not_eval_samples": 0,
|
| 237 |
+
"macro_mean_score": 0.3502244283768534,
|
| 238 |
+
"micro_mean_score": 0.3456783051732046
|
|
|
|
| 239 |
},
|
| 240 |
"core_cot": {
|
| 241 |
"num_eval_tasks": 440,
|
| 242 |
"num_eval_samples": 6539,
|
| 243 |
"num_not_eval_samples": 0,
|
| 244 |
+
"macro_mean_score": 0.3562710424410931,
|
| 245 |
+
"micro_mean_score": 0.35129859801162616
|
|
|
|
| 246 |
},
|
| 247 |
"open": {
|
| 248 |
"num_eval_tasks": 65,
|
| 249 |
"num_eval_samples": 1163,
|
|
|
|
| 250 |
"macro_mean_score": 0.5192997443033639,
|
| 251 |
"micro_mean_score": 0.5421324161650903
|
| 252 |
},
|
| 253 |
+
"overall_score": 0.3772549347599992
|
| 254 |
},
|
| 255 |
"InternVL2_8B": {
|
| 256 |
"core_noncot": {
|
| 257 |
"num_eval_tasks": 440,
|
| 258 |
"num_eval_samples": 6539,
|
| 259 |
"num_not_eval_samples": 0,
|
| 260 |
+
"macro_mean_score": 0.25956581776451815,
|
| 261 |
+
"micro_mean_score": 0.2546984460483302
|
|
|
|
| 262 |
},
|
| 263 |
"core_cot": {
|
| 264 |
"num_eval_tasks": 440,
|
| 265 |
"num_eval_samples": 6539,
|
| 266 |
"num_not_eval_samples": 0,
|
| 267 |
+
"macro_mean_score": 0.24090301358258295,
|
| 268 |
+
"micro_mean_score": 0.23819084111520938
|
|
|
|
| 269 |
},
|
| 270 |
"open": {
|
| 271 |
"num_eval_tasks": 65,
|
| 272 |
"num_eval_samples": 1165,
|
|
|
|
| 273 |
"macro_mean_score": 0.3978571701460552,
|
| 274 |
"micro_mean_score": 0.4108583690987125
|
| 275 |
},
|
| 276 |
+
"overall_score": 0.2773656948037259
|
| 277 |
},
|
| 278 |
"MiniCPM_v2.6": {
|
| 279 |
"core_noncot": {
|
| 280 |
"num_eval_tasks": 440,
|
| 281 |
"num_eval_samples": 6539,
|
| 282 |
"num_not_eval_samples": 0,
|
| 283 |
+
"macro_mean_score": 0.2287645706203155,
|
| 284 |
+
"micro_mean_score": 0.2249087742955901
|
|
|
|
| 285 |
},
|
| 286 |
"core_cot": {
|
| 287 |
"num_eval_tasks": 440,
|
| 288 |
"num_eval_samples": 6539,
|
| 289 |
"num_not_eval_samples": 0,
|
| 290 |
+
"macro_mean_score": 0.22955895202146906,
|
| 291 |
+
"micro_mean_score": 0.22560399396899078
|
|
|
|
| 292 |
},
|
| 293 |
"open": {
|
| 294 |
"num_eval_tasks": 65,
|
| 295 |
"num_eval_samples": 1163,
|
|
|
|
| 296 |
"macro_mean_score": 0.41728623355613875,
|
| 297 |
"micro_mean_score": 0.43452278589853827
|
| 298 |
},
|
| 299 |
+
"overall_score": 0.2537218694467236
|
| 300 |
},
|
| 301 |
"Phi-3.5-vision": {
|
| 302 |
"core_noncot": {
|
| 303 |
"num_eval_tasks": 440,
|
| 304 |
"num_eval_samples": 6539,
|
| 305 |
"num_not_eval_samples": 0,
|
| 306 |
+
"macro_mean_score": 0.23271251159409778,
|
| 307 |
+
"micro_mean_score": 0.2296262323791101
|
|
|
|
| 308 |
},
|
| 309 |
"core_cot": {
|
| 310 |
"num_eval_tasks": 440,
|
| 311 |
"num_eval_samples": 6539,
|
| 312 |
"num_not_eval_samples": 0,
|
| 313 |
+
"macro_mean_score": 0.22995297916629392,
|
| 314 |
+
"micro_mean_score": 0.22708502951025372
|
|
|
|
| 315 |
},
|
| 316 |
"open": {
|
| 317 |
"num_eval_tasks": 65,
|
| 318 |
"num_eval_samples": 1163,
|
|
|
|
| 319 |
"macro_mean_score": 0.3947914647737769,
|
| 320 |
"micro_mean_score": 0.42459157351676696
|
| 321 |
},
|
| 322 |
+
"overall_score": 0.25357415903306635
|
| 323 |
},
|
| 324 |
"Pixtral_12B": {
|
| 325 |
"core_noncot": {
|
| 326 |
"num_eval_tasks": 440,
|
| 327 |
"num_eval_samples": 6539,
|
| 328 |
"num_not_eval_samples": 0,
|
| 329 |
+
"macro_mean_score": 0.31905695620134694,
|
| 330 |
+
"micro_mean_score": 0.31556607913724777
|
|
|
|
| 331 |
},
|
| 332 |
"core_cot": {
|
| 333 |
"num_eval_tasks": 440,
|
| 334 |
"num_eval_samples": 6539,
|
| 335 |
"num_not_eval_samples": 0,
|
| 336 |
+
"macro_mean_score": 0.31362045151669854,
|
| 337 |
+
"micro_mean_score": 0.3100986209078182
|
|
|
|
| 338 |
},
|
| 339 |
"open": {
|
| 340 |
"num_eval_tasks": 65,
|
| 341 |
"num_eval_samples": 1163,
|
|
|
|
| 342 |
"macro_mean_score": 0.4566234428542061,
|
| 343 |
"micro_mean_score": 0.4870593293207223
|
| 344 |
},
|
| 345 |
+
"overall_score": 0.33676353369131895
|
| 346 |
},
|
| 347 |
"Llama_3_2_11B": {
|
| 348 |
"core_noncot": {
|
| 349 |
"num_eval_tasks": 440,
|
| 350 |
"num_eval_samples": 6539,
|
| 351 |
"num_not_eval_samples": 0,
|
|
|
|
| 352 |
"macro_mean_score": 0.10044261716549671,
|
| 353 |
"micro_mean_score": 0.09980638766828835
|
| 354 |
},
|
|
|
|
| 356 |
"num_eval_tasks": 440,
|
| 357 |
"num_eval_samples": 6539,
|
| 358 |
"num_not_eval_samples": 0,
|
| 359 |
+
"macro_mean_score": 0.15999641916771298,
|
| 360 |
+
"micro_mean_score": 0.15809331016967038
|
|
|
|
| 361 |
},
|
| 362 |
"open": {
|
| 363 |
"num_eval_tasks": 65,
|
| 364 |
"num_eval_samples": 1163,
|
|
|
|
| 365 |
"macro_mean_score": 0.3173342406187366,
|
| 366 |
"micro_mean_score": 0.3487962166809973
|
| 367 |
},
|
| 368 |
+
"overall_score": 0.1802478219287358
|
| 369 |
},
|
| 370 |
"Idefics3": {
|
| 371 |
"core_noncot": {
|
| 372 |
"num_eval_tasks": 440,
|
| 373 |
"num_eval_samples": 6539,
|
| 374 |
"num_not_eval_samples": 0,
|
|
|
|
| 375 |
"macro_mean_score": 0.11118980301103833,
|
| 376 |
"micro_mean_score": 0.11201785633274061
|
| 377 |
},
|
|
|
|
| 379 |
"num_eval_tasks": 440,
|
| 380 |
"num_eval_samples": 6539,
|
| 381 |
"num_not_eval_samples": 0,
|
|
|
|
| 382 |
"macro_mean_score": 0.08956972487602757,
|
| 383 |
"micro_mean_score": 0.08982225274252693
|
| 384 |
},
|
| 385 |
"open": {
|
| 386 |
"num_eval_tasks": 65,
|
| 387 |
"num_eval_samples": 1163,
|
|
|
|
| 388 |
"macro_mean_score": 0.3210866162255635,
|
| 389 |
"micro_mean_score": 0.35649183147033553
|
| 390 |
},
|
|
|
|
| 395 |
"num_eval_tasks": 440,
|
| 396 |
"num_eval_samples": 6539,
|
| 397 |
"num_not_eval_samples": 0,
|
|
|
|
| 398 |
"macro_mean_score": 0.30485930718699694,
|
| 399 |
"micro_mean_score": 0.3016713629035311
|
| 400 |
},
|
|
|
|
| 402 |
"num_eval_tasks": 440,
|
| 403 |
"num_eval_samples": 6539,
|
| 404 |
"num_not_eval_samples": 0,
|
|
|
|
| 405 |
"macro_mean_score": 0.289073788209904,
|
| 406 |
"micro_mean_score": 0.2859007507765791
|
| 407 |
},
|
| 408 |
"open": {
|
| 409 |
"num_eval_tasks": 65,
|
| 410 |
"num_eval_samples": 1163,
|
|
|
|
| 411 |
"macro_mean_score": 0.5103725263180767,
|
| 412 |
"micro_mean_score": 0.5349957007738607
|
| 413 |
},
|
|
|
|
| 418 |
"num_eval_tasks": 440,
|
| 419 |
"num_eval_samples": 6539,
|
| 420 |
"num_not_eval_samples": 0,
|
|
|
|
| 421 |
"macro_mean_score": 0.2420528895703979,
|
| 422 |
"micro_mean_score": 0.23838419989257642
|
| 423 |
},
|
|
|
|
| 425 |
"num_eval_tasks": 440,
|
| 426 |
"num_eval_samples": 6539,
|
| 427 |
"num_not_eval_samples": 0,
|
|
|
|
| 428 |
"macro_mean_score": 0.21589726765847422,
|
| 429 |
"micro_mean_score": 0.21406043849932396
|
| 430 |
},
|
| 431 |
"open": {
|
| 432 |
"num_eval_tasks": 65,
|
| 433 |
"num_eval_samples": 1163,
|
|
|
|
| 434 |
"macro_mean_score": 0.3478114310231307,
|
| 435 |
"micro_mean_score": 0.3947549441100602
|
| 436 |
},
|
|
|
|
| 441 |
"num_eval_tasks": 440,
|
| 442 |
"num_eval_samples": 6539,
|
| 443 |
"num_not_eval_samples": 0,
|
|
|
|
| 444 |
"macro_mean_score": 0.09089701489596874,
|
| 445 |
"micro_mean_score": 0.09036328295381871
|
| 446 |
},
|
|
|
|
| 448 |
"num_eval_tasks": 440,
|
| 449 |
"num_eval_samples": 6539,
|
| 450 |
"num_not_eval_samples": 0,
|
|
|
|
| 451 |
"macro_mean_score": 0.13141974398938763,
|
| 452 |
"micro_mean_score": 0.13063500716262516
|
| 453 |
},
|
| 454 |
"open": {
|
| 455 |
"num_eval_tasks": 65,
|
| 456 |
"num_eval_samples": 1163,
|
|
|
|
| 457 |
"macro_mean_score": 0.23864417043743646,
|
| 458 |
"micro_mean_score": 0.24901117798796224
|
| 459 |
},
|
|
|
|
| 464 |
"num_eval_tasks": 440,
|
| 465 |
"num_eval_samples": 6539,
|
| 466 |
"num_not_eval_samples": 0,
|
|
|
|
| 467 |
"macro_mean_score": 0.16448220309703876,
|
| 468 |
"micro_mean_score": 0.1610710186451323
|
| 469 |
},
|
|
|
|
| 471 |
"num_eval_tasks": 440,
|
| 472 |
"num_eval_samples": 6539,
|
| 473 |
"num_not_eval_samples": 0,
|
|
|
|
| 474 |
"macro_mean_score": 0.20877163406364055,
|
| 475 |
"micro_mean_score": 0.20561526268932287
|
| 476 |
},
|
| 477 |
"open": {
|
| 478 |
"num_eval_tasks": 65,
|
| 479 |
"num_eval_samples": 1163,
|
|
|
|
| 480 |
"macro_mean_score": 0.3154302566225611,
|
| 481 |
"micro_mean_score": 0.33856405846947557
|
| 482 |
},
|
|
|
|
| 487 |
"num_eval_tasks": 440,
|
| 488 |
"num_eval_samples": 6539,
|
| 489 |
"num_not_eval_samples": 0,
|
|
|
|
| 490 |
"macro_mean_score": 0.16317824309838627,
|
| 491 |
"micro_mean_score": 0.16198837245148487
|
| 492 |
},
|
|
|
|
| 494 |
"num_eval_tasks": 440,
|
| 495 |
"num_eval_samples": 6539,
|
| 496 |
"num_not_eval_samples": 0,
|
|
|
|
| 497 |
"macro_mean_score": 0.159970161379836,
|
| 498 |
"micro_mean_score": 0.15844711671722148
|
| 499 |
},
|
| 500 |
"open": {
|
| 501 |
"num_eval_tasks": 65,
|
| 502 |
"num_eval_samples": 1163,
|
|
|
|
| 503 |
"macro_mean_score": 0.24567572098570653,
|
| 504 |
"micro_mean_score": 0.2704213241616509
|
| 505 |
},
|
static/eval_results/{Core_SI → SI}/all_model_keywords_stats.json
RENAMED
|
The diff for this file is too large to render.
See raw diff
|
|
|
static/eval_results/SI/all_summary.json
ADDED
|
@@ -0,0 +1,471 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"Aquila_VL_2B": {
|
| 3 |
+
"core": {
|
| 4 |
+
"num_eval_tasks": 273,
|
| 5 |
+
"num_eval_samples": 4116,
|
| 6 |
+
"num_not_eval_samples": 0,
|
| 7 |
+
"macro_mean_score": 0.20770364903712493,
|
| 8 |
+
"micro_mean_score": 0.20333142638522636,
|
| 9 |
+
"missing_tasks": []
|
| 10 |
+
},
|
| 11 |
+
"open": {
|
| 12 |
+
"num_eval_tasks": 42,
|
| 13 |
+
"num_eval_samples": 813,
|
| 14 |
+
"num_not_eval_samples": 0,
|
| 15 |
+
"macro_mean_score": 0.31474202723571276,
|
| 16 |
+
"micro_mean_score": 0.3326568265682657,
|
| 17 |
+
"missing_tasks": []
|
| 18 |
+
},
|
| 19 |
+
"overall_score": 0.22197543279693666
|
| 20 |
+
},
|
| 21 |
+
"Aria": {
|
| 22 |
+
"core": {
|
| 23 |
+
"num_eval_tasks": 273,
|
| 24 |
+
"num_eval_samples": 4116,
|
| 25 |
+
"num_not_eval_samples": 0,
|
| 26 |
+
"macro_mean_score": 0.3178882776147889,
|
| 27 |
+
"micro_mean_score": 0.3101511832828904,
|
| 28 |
+
"missing_tasks": []
|
| 29 |
+
},
|
| 30 |
+
"open": {
|
| 31 |
+
"num_eval_tasks": 42,
|
| 32 |
+
"num_eval_samples": 813,
|
| 33 |
+
"num_not_eval_samples": 0,
|
| 34 |
+
"macro_mean_score": 0.5137437248005172,
|
| 35 |
+
"micro_mean_score": 0.5472939729397295,
|
| 36 |
+
"missing_tasks": []
|
| 37 |
+
},
|
| 38 |
+
"overall_score": 0.34400233723955265
|
| 39 |
+
},
|
| 40 |
+
"Claude_3.5": {
|
| 41 |
+
"core": {
|
| 42 |
+
"num_eval_tasks": 273,
|
| 43 |
+
"num_eval_samples": 4116,
|
| 44 |
+
"num_not_eval_samples": 0,
|
| 45 |
+
"macro_mean_score": 0.520276385877485,
|
| 46 |
+
"micro_mean_score": 0.5148202137998056
|
| 47 |
+
},
|
| 48 |
+
"open": {
|
| 49 |
+
"num_eval_tasks": 42,
|
| 50 |
+
"num_eval_samples": 813,
|
| 51 |
+
"num_not_eval_samples": 0,
|
| 52 |
+
"macro_mean_score": 0.6479684260295507,
|
| 53 |
+
"micro_mean_score": 0.6801968019680197
|
| 54 |
+
},
|
| 55 |
+
"overall_score": 0.5373019912310938
|
| 56 |
+
},
|
| 57 |
+
"Claude_3.5_new": {
|
| 58 |
+
"core": {
|
| 59 |
+
"num_eval_tasks": 273,
|
| 60 |
+
"num_eval_samples": 4116,
|
| 61 |
+
"num_not_eval_samples": 0,
|
| 62 |
+
"macro_mean_score": 0.5462752278980763,
|
| 63 |
+
"micro_mean_score": 0.5417881438289601
|
| 64 |
+
},
|
| 65 |
+
"open": {
|
| 66 |
+
"num_eval_tasks": 42,
|
| 67 |
+
"num_eval_samples": 813,
|
| 68 |
+
"num_not_eval_samples": 0,
|
| 69 |
+
"macro_mean_score": 0.6764020657053476,
|
| 70 |
+
"micro_mean_score": 0.6924969249692496
|
| 71 |
+
},
|
| 72 |
+
"overall_score": 0.5636254729390457
|
| 73 |
+
},
|
| 74 |
+
"GPT_4o": {
|
| 75 |
+
"core": {
|
| 76 |
+
"num_eval_tasks": 273,
|
| 77 |
+
"num_eval_samples": 4116,
|
| 78 |
+
"num_not_eval_samples": 0,
|
| 79 |
+
"macro_mean_score": 0.5529953662872719,
|
| 80 |
+
"micro_mean_score": 0.5483479105928085
|
| 81 |
+
},
|
| 82 |
+
"open": {
|
| 83 |
+
"num_eval_tasks": 42,
|
| 84 |
+
"num_eval_samples": 813,
|
| 85 |
+
"num_not_eval_samples": 0,
|
| 86 |
+
"macro_mean_score": 0.6600228904804206,
|
| 87 |
+
"micro_mean_score": 0.6801968019680197
|
| 88 |
+
},
|
| 89 |
+
"overall_score": 0.5672657028463584
|
| 90 |
+
},
|
| 91 |
+
"GPT_4o_mini": {
|
| 92 |
+
"core": {
|
| 93 |
+
"num_eval_tasks": 273,
|
| 94 |
+
"num_eval_samples": 4116,
|
| 95 |
+
"num_not_eval_samples": 0,
|
| 96 |
+
"macro_mean_score": 0.44285970964797233,
|
| 97 |
+
"micro_mean_score": 0.43756073858114675
|
| 98 |
+
},
|
| 99 |
+
"open": {
|
| 100 |
+
"num_eval_tasks": 42,
|
| 101 |
+
"num_eval_samples": 813,
|
| 102 |
+
"num_not_eval_samples": 0,
|
| 103 |
+
"macro_mean_score": 0.595574663769726,
|
| 104 |
+
"micro_mean_score": 0.6334563345633456
|
| 105 |
+
},
|
| 106 |
+
"overall_score": 0.46322170353087283
|
| 107 |
+
},
|
| 108 |
+
"Gemini_1.5_flash_002": {
|
| 109 |
+
"core": {
|
| 110 |
+
"num_eval_tasks": 273,
|
| 111 |
+
"num_eval_samples": 4116,
|
| 112 |
+
"num_not_eval_samples": 0,
|
| 113 |
+
"macro_mean_score": 0.42188460865574384,
|
| 114 |
+
"micro_mean_score": 0.413508260447036
|
| 115 |
+
},
|
| 116 |
+
"open": {
|
| 117 |
+
"num_eval_tasks": 42,
|
| 118 |
+
"num_eval_samples": 813,
|
| 119 |
+
"num_not_eval_samples": 0,
|
| 120 |
+
"macro_mean_score": 0.5787083135236054,
|
| 121 |
+
"micro_mean_score": 0.6186961869618696
|
| 122 |
+
},
|
| 123 |
+
"overall_score": 0.44279443597145873
|
| 124 |
+
},
|
| 125 |
+
"Gemini_1.5_pro_002": {
|
| 126 |
+
"core": {
|
| 127 |
+
"num_eval_tasks": 273,
|
| 128 |
+
"num_eval_samples": 4116,
|
| 129 |
+
"num_not_eval_samples": 0,
|
| 130 |
+
"macro_mean_score": 0.4914311038229404,
|
| 131 |
+
"micro_mean_score": 0.48323615160349853
|
| 132 |
+
},
|
| 133 |
+
"open": {
|
| 134 |
+
"num_eval_tasks": 42,
|
| 135 |
+
"num_eval_samples": 813,
|
| 136 |
+
"num_not_eval_samples": 0,
|
| 137 |
+
"macro_mean_score": 0.5814975405131552,
|
| 138 |
+
"micro_mean_score": 0.6174661746617466
|
| 139 |
+
},
|
| 140 |
+
"overall_score": 0.5034399620483024
|
| 141 |
+
},
|
| 142 |
+
"Idefics3": {
|
| 143 |
+
"core": {
|
| 144 |
+
"num_eval_tasks": 273,
|
| 145 |
+
"num_eval_samples": 4116,
|
| 146 |
+
"num_not_eval_samples": 0,
|
| 147 |
+
"macro_mean_score": 0.08941182847569326,
|
| 148 |
+
"micro_mean_score": 0.08779475233900695,
|
| 149 |
+
"missing_tasks": []
|
| 150 |
+
},
|
| 151 |
+
"open": {
|
| 152 |
+
"num_eval_tasks": 42,
|
| 153 |
+
"num_eval_samples": 813,
|
| 154 |
+
"num_not_eval_samples": 0,
|
| 155 |
+
"macro_mean_score": 0.3231434267517844,
|
| 156 |
+
"micro_mean_score": 0.3618081180811809,
|
| 157 |
+
"missing_tasks": []
|
| 158 |
+
},
|
| 159 |
+
"overall_score": 0.12057604157917208
|
| 160 |
+
},
|
| 161 |
+
"InternVL2_2B": {
|
| 162 |
+
"core": {
|
| 163 |
+
"num_eval_tasks": 273,
|
| 164 |
+
"num_eval_samples": 4116,
|
| 165 |
+
"num_not_eval_samples": 0,
|
| 166 |
+
"macro_mean_score": 0.12069001041308772,
|
| 167 |
+
"micro_mean_score": 0.11842605219090299,
|
| 168 |
+
"missing_tasks": []
|
| 169 |
+
},
|
| 170 |
+
"open": {
|
| 171 |
+
"num_eval_tasks": 42,
|
| 172 |
+
"num_eval_samples": 813,
|
| 173 |
+
"num_not_eval_samples": 0,
|
| 174 |
+
"macro_mean_score": 0.28522459992910454,
|
| 175 |
+
"micro_mean_score": 0.28886838868388687,
|
| 176 |
+
"missing_tasks": []
|
| 177 |
+
},
|
| 178 |
+
"overall_score": 0.14262795568189
|
| 179 |
+
},
|
| 180 |
+
"InternVL2_76B": {
|
| 181 |
+
"core": {
|
| 182 |
+
"num_eval_tasks": 273,
|
| 183 |
+
"num_eval_samples": 4116,
|
| 184 |
+
"num_not_eval_samples": 0,
|
| 185 |
+
"macro_mean_score": 0.3998616568018755,
|
| 186 |
+
"micro_mean_score": 0.39149064302628933,
|
| 187 |
+
"missing_tasks": []
|
| 188 |
+
},
|
| 189 |
+
"open": {
|
| 190 |
+
"num_eval_tasks": 42,
|
| 191 |
+
"num_eval_samples": 813,
|
| 192 |
+
"num_not_eval_samples": 0,
|
| 193 |
+
"macro_mean_score": 0.554748737158244,
|
| 194 |
+
"micro_mean_score": 0.5800738007380073,
|
| 195 |
+
"missing_tasks": []
|
| 196 |
+
},
|
| 197 |
+
"overall_score": 0.42051326751605805
|
| 198 |
+
},
|
| 199 |
+
"InternVL2_8B": {
|
| 200 |
+
"core": {
|
| 201 |
+
"num_eval_tasks": 273,
|
| 202 |
+
"num_eval_samples": 4116,
|
| 203 |
+
"num_not_eval_samples": 0,
|
| 204 |
+
"macro_mean_score": 0.27650612401825575,
|
| 205 |
+
"micro_mean_score": 0.27119471729837735,
|
| 206 |
+
"missing_tasks": []
|
| 207 |
+
},
|
| 208 |
+
"open": {
|
| 209 |
+
"num_eval_tasks": 42,
|
| 210 |
+
"num_eval_samples": 813,
|
| 211 |
+
"num_not_eval_samples": 0,
|
| 212 |
+
"macro_mean_score": 0.39388373890935635,
|
| 213 |
+
"micro_mean_score": 0.4045510455104551,
|
| 214 |
+
"missing_tasks": []
|
| 215 |
+
},
|
| 216 |
+
"overall_score": 0.29215647267040246
|
| 217 |
+
},
|
| 218 |
+
"Llama_3_2_11B": {
|
| 219 |
+
"core": {
|
| 220 |
+
"num_eval_tasks": 273,
|
| 221 |
+
"num_eval_samples": 4116,
|
| 222 |
+
"num_not_eval_samples": 0,
|
| 223 |
+
"macro_mean_score": 0.20789144960796493,
|
| 224 |
+
"micro_mean_score": 0.20163641703273802,
|
| 225 |
+
"missing_tasks": []
|
| 226 |
+
},
|
| 227 |
+
"open": {
|
| 228 |
+
"num_eval_tasks": 42,
|
| 229 |
+
"num_eval_samples": 813,
|
| 230 |
+
"num_not_eval_samples": 0,
|
| 231 |
+
"macro_mean_score": 0.3861125858565788,
|
| 232 |
+
"micro_mean_score": 0.4130381303813038,
|
| 233 |
+
"missing_tasks": []
|
| 234 |
+
},
|
| 235 |
+
"overall_score": 0.2316542677744468
|
| 236 |
+
},
|
| 237 |
+
"MiniCPM_v2.6": {
|
| 238 |
+
"core": {
|
| 239 |
+
"num_eval_tasks": 273,
|
| 240 |
+
"num_eval_samples": 4116,
|
| 241 |
+
"num_not_eval_samples": 0,
|
| 242 |
+
"macro_mean_score": 0.23230765810722817,
|
| 243 |
+
"micro_mean_score": 0.22684118052665975,
|
| 244 |
+
"missing_tasks": []
|
| 245 |
+
},
|
| 246 |
+
"open": {
|
| 247 |
+
"num_eval_tasks": 42,
|
| 248 |
+
"num_eval_samples": 813,
|
| 249 |
+
"num_not_eval_samples": 0,
|
| 250 |
+
"macro_mean_score": 0.4360655066213874,
|
| 251 |
+
"micro_mean_score": 0.4588560885608856,
|
| 252 |
+
"missing_tasks": []
|
| 253 |
+
},
|
| 254 |
+
"overall_score": 0.2594753712424494
|
| 255 |
+
},
|
| 256 |
+
"Molmo_72B": {
|
| 257 |
+
"core": {
|
| 258 |
+
"num_eval_tasks": 270,
|
| 259 |
+
"num_eval_samples": 4073,
|
| 260 |
+
"num_not_eval_samples": 0,
|
| 261 |
+
"macro_mean_score": 0.36480000609384927,
|
| 262 |
+
"micro_mean_score": 0.36205779758110807,
|
| 263 |
+
"missing_tasks": [
|
| 264 |
+
"MMSoc_Misinformation_PolitiFact",
|
| 265 |
+
"table_understanding",
|
| 266 |
+
"planning_screenshot_termes"
|
| 267 |
+
]
|
| 268 |
+
},
|
| 269 |
+
"open": {
|
| 270 |
+
"num_eval_tasks": 42,
|
| 271 |
+
"num_eval_samples": 813,
|
| 272 |
+
"num_not_eval_samples": 0,
|
| 273 |
+
"macro_mean_score": 0.4465682063915481,
|
| 274 |
+
"micro_mean_score": 0.4850553505535054,
|
| 275 |
+
"missing_tasks": []
|
| 276 |
+
},
|
| 277 |
+
"overall_score": 0.3758072638262318
|
| 278 |
+
},
|
| 279 |
+
"Molmo_7B_D": {
|
| 280 |
+
"core": {
|
| 281 |
+
"num_eval_tasks": 272,
|
| 282 |
+
"num_eval_samples": 4102,
|
| 283 |
+
"num_not_eval_samples": 0,
|
| 284 |
+
"macro_mean_score": 0.2098088446992518,
|
| 285 |
+
"micro_mean_score": 0.20550929661464645,
|
| 286 |
+
"missing_tasks": [
|
| 287 |
+
"MMSoc_Misinformation_PolitiFact"
|
| 288 |
+
]
|
| 289 |
+
},
|
| 290 |
+
"open": {
|
| 291 |
+
"num_eval_tasks": 42,
|
| 292 |
+
"num_eval_samples": 813,
|
| 293 |
+
"num_not_eval_samples": 0,
|
| 294 |
+
"macro_mean_score": 0.35697926179118733,
|
| 295 |
+
"micro_mean_score": 0.38936039360393604,
|
| 296 |
+
"missing_tasks": []
|
| 297 |
+
},
|
| 298 |
+
"overall_score": 0.22949405972428777
|
| 299 |
+
},
|
| 300 |
+
"NVLM": {
|
| 301 |
+
"core": {
|
| 302 |
+
"num_eval_tasks": 273,
|
| 303 |
+
"num_eval_samples": 4116,
|
| 304 |
+
"num_not_eval_samples": 0,
|
| 305 |
+
"macro_mean_score": 0.32989872890926025,
|
| 306 |
+
"micro_mean_score": 0.32315683713111915,
|
| 307 |
+
"missing_tasks": []
|
| 308 |
+
},
|
| 309 |
+
"open": {
|
| 310 |
+
"num_eval_tasks": 42,
|
| 311 |
+
"num_eval_samples": 813,
|
| 312 |
+
"num_not_eval_samples": 0,
|
| 313 |
+
"macro_mean_score": 0.4469349818134809,
|
| 314 |
+
"micro_mean_score": 0.4881303813038132,
|
| 315 |
+
"missing_tasks": []
|
| 316 |
+
},
|
| 317 |
+
"overall_score": 0.34550356262982296
|
| 318 |
+
},
|
| 319 |
+
"POINTS_7B": {
|
| 320 |
+
"core": {
|
| 321 |
+
"num_eval_tasks": 273,
|
| 322 |
+
"num_eval_samples": 4116,
|
| 323 |
+
"num_not_eval_samples": 0,
|
| 324 |
+
"macro_mean_score": 0.25511317681632334,
|
| 325 |
+
"micro_mean_score": 0.24927711632415062,
|
| 326 |
+
"missing_tasks": []
|
| 327 |
+
},
|
| 328 |
+
"open": {
|
| 329 |
+
"num_eval_tasks": 42,
|
| 330 |
+
"num_eval_samples": 813,
|
| 331 |
+
"num_not_eval_samples": 0,
|
| 332 |
+
"macro_mean_score": 0.30315625179016,
|
| 333 |
+
"micro_mean_score": 0.3313653136531366,
|
| 334 |
+
"missing_tasks": []
|
| 335 |
+
},
|
| 336 |
+
"overall_score": 0.26151892014616823
|
| 337 |
+
},
|
| 338 |
+
"Phi-3.5-vision": {
|
| 339 |
+
"core": {
|
| 340 |
+
"num_eval_tasks": 273,
|
| 341 |
+
"num_eval_samples": 4116,
|
| 342 |
+
"num_not_eval_samples": 0,
|
| 343 |
+
"macro_mean_score": 0.2561274958722834,
|
| 344 |
+
"micro_mean_score": 0.2504214576875906,
|
| 345 |
+
"missing_tasks": []
|
| 346 |
+
},
|
| 347 |
+
"open": {
|
| 348 |
+
"num_eval_tasks": 42,
|
| 349 |
+
"num_eval_samples": 813,
|
| 350 |
+
"num_not_eval_samples": 0,
|
| 351 |
+
"macro_mean_score": 0.4272267419054576,
|
| 352 |
+
"micro_mean_score": 0.445879458794588,
|
| 353 |
+
"missing_tasks": []
|
| 354 |
+
},
|
| 355 |
+
"overall_score": 0.2789407286767066
|
| 356 |
+
},
|
| 357 |
+
"Pixtral_12B": {
|
| 358 |
+
"core": {
|
| 359 |
+
"num_eval_tasks": 273,
|
| 360 |
+
"num_eval_samples": 4116,
|
| 361 |
+
"num_not_eval_samples": 0,
|
| 362 |
+
"macro_mean_score": 0.3436942439614412,
|
| 363 |
+
"micro_mean_score": 0.3373564384613738,
|
| 364 |
+
"missing_tasks": []
|
| 365 |
+
},
|
| 366 |
+
"open": {
|
| 367 |
+
"num_eval_tasks": 42,
|
| 368 |
+
"num_eval_samples": 813,
|
| 369 |
+
"num_not_eval_samples": 0,
|
| 370 |
+
"macro_mean_score": 0.4417271955536318,
|
| 371 |
+
"micro_mean_score": 0.4845633456334564,
|
| 372 |
+
"missing_tasks": []
|
| 373 |
+
},
|
| 374 |
+
"overall_score": 0.3567653041737333
|
| 375 |
+
},
|
| 376 |
+
"Qwen2_VL_2B": {
|
| 377 |
+
"core": {
|
| 378 |
+
"num_eval_tasks": 273,
|
| 379 |
+
"num_eval_samples": 4116,
|
| 380 |
+
"num_not_eval_samples": 0,
|
| 381 |
+
"macro_mean_score": 0.22787906973244856,
|
| 382 |
+
"micro_mean_score": 0.2234748515064842,
|
| 383 |
+
"missing_tasks": []
|
| 384 |
+
},
|
| 385 |
+
"open": {
|
| 386 |
+
"num_eval_tasks": 42,
|
| 387 |
+
"num_eval_samples": 813,
|
| 388 |
+
"num_not_eval_samples": 0,
|
| 389 |
+
"macro_mean_score": 0.3509364634962041,
|
| 390 |
+
"micro_mean_score": 0.3768757687576875,
|
| 391 |
+
"missing_tasks": []
|
| 392 |
+
},
|
| 393 |
+
"overall_score": 0.24428672223428263
|
| 394 |
+
},
|
| 395 |
+
"Qwen2_VL_72B": {
|
| 396 |
+
"core": {
|
| 397 |
+
"num_eval_tasks": 273,
|
| 398 |
+
"num_eval_samples": 4116,
|
| 399 |
+
"num_not_eval_samples": 0,
|
| 400 |
+
"macro_mean_score": 0.4730536307784527,
|
| 401 |
+
"micro_mean_score": 0.4659830915476831,
|
| 402 |
+
"missing_tasks": []
|
| 403 |
+
},
|
| 404 |
+
"open": {
|
| 405 |
+
"num_eval_tasks": 42,
|
| 406 |
+
"num_eval_samples": 813,
|
| 407 |
+
"num_not_eval_samples": 0,
|
| 408 |
+
"macro_mean_score": 0.5510079982505317,
|
| 409 |
+
"micro_mean_score": 0.5826568265682657,
|
| 410 |
+
"missing_tasks": []
|
| 411 |
+
},
|
| 412 |
+
"overall_score": 0.48344754644139654
|
| 413 |
+
},
|
| 414 |
+
"Qwen2_VL_7B": {
|
| 415 |
+
"core": {
|
| 416 |
+
"num_eval_tasks": 273,
|
| 417 |
+
"num_eval_samples": 4116,
|
| 418 |
+
"num_not_eval_samples": 0,
|
| 419 |
+
"macro_mean_score": 0.3538656561495699,
|
| 420 |
+
"micro_mean_score": 0.34581250459157137,
|
| 421 |
+
"missing_tasks": []
|
| 422 |
+
},
|
| 423 |
+
"open": {
|
| 424 |
+
"num_eval_tasks": 42,
|
| 425 |
+
"num_eval_samples": 813,
|
| 426 |
+
"num_not_eval_samples": 0,
|
| 427 |
+
"macro_mean_score": 0.4517429592549692,
|
| 428 |
+
"micro_mean_score": 0.4730012300123002,
|
| 429 |
+
"missing_tasks": []
|
| 430 |
+
},
|
| 431 |
+
"overall_score": 0.3669159632302898
|
| 432 |
+
},
|
| 433 |
+
"llava_onevision_72B": {
|
| 434 |
+
"core": {
|
| 435 |
+
"num_eval_tasks": 273,
|
| 436 |
+
"num_eval_samples": 4116,
|
| 437 |
+
"num_not_eval_samples": 0,
|
| 438 |
+
"macro_mean_score": 0.312618242621264,
|
| 439 |
+
"micro_mean_score": 0.3098623876487132,
|
| 440 |
+
"missing_tasks": []
|
| 441 |
+
},
|
| 442 |
+
"open": {
|
| 443 |
+
"num_eval_tasks": 42,
|
| 444 |
+
"num_eval_samples": 813,
|
| 445 |
+
"num_not_eval_samples": 0,
|
| 446 |
+
"macro_mean_score": 0.4425822460912829,
|
| 447 |
+
"micro_mean_score": 0.47539975399754,
|
| 448 |
+
"missing_tasks": []
|
| 449 |
+
},
|
| 450 |
+
"overall_score": 0.32994677641726655
|
| 451 |
+
},
|
| 452 |
+
"llava_onevision_7B": {
|
| 453 |
+
"core": {
|
| 454 |
+
"num_eval_tasks": 273,
|
| 455 |
+
"num_eval_samples": 4116,
|
| 456 |
+
"num_not_eval_samples": 0,
|
| 457 |
+
"macro_mean_score": 0.23683339637631812,
|
| 458 |
+
"micro_mean_score": 0.23283041278687175,
|
| 459 |
+
"missing_tasks": []
|
| 460 |
+
},
|
| 461 |
+
"open": {
|
| 462 |
+
"num_eval_tasks": 42,
|
| 463 |
+
"num_eval_samples": 813,
|
| 464 |
+
"num_not_eval_samples": 0,
|
| 465 |
+
"macro_mean_score": 0.3871602360316429,
|
| 466 |
+
"micro_mean_score": 0.4113161131611316,
|
| 467 |
+
"missing_tasks": []
|
| 468 |
+
},
|
| 469 |
+
"overall_score": 0.25687697499702805
|
| 470 |
+
}
|
| 471 |
+
}
|
utils.py
CHANGED
|
@@ -11,7 +11,7 @@ MODEL_NAME_MAP = {
|
|
| 11 |
"InternVL2_76B": "InternVL2-Llama3-76B",
|
| 12 |
"Qwen2_VL_72B": "Qwen2-VL-72B",
|
| 13 |
"llava_onevision_72B": "Llava-OneVision-72B",
|
| 14 |
-
"NVLM": "NVLM-72B",
|
| 15 |
"GPT_4o_mini": "GPT-4o mini",
|
| 16 |
"Gemini_1.5_flash_002": "Gemini-1.5-Flash-002",
|
| 17 |
"Pixtral_12B": "Pixtral 12B",
|
|
@@ -83,6 +83,34 @@ KEYWORD_NAME_MAP = {
|
|
| 83 |
"video": "Video",
|
| 84 |
}
|
| 85 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 86 |
class BaseDataLoader:
|
| 87 |
# Define the base MODEL_GROUPS structure
|
| 88 |
BASE_MODEL_GROUPS = {
|
|
@@ -183,10 +211,10 @@ class DefaultDataLoader(BaseDataLoader):
|
|
| 183 |
core_noncot_score = summary["core_noncot"]["macro_mean_score"]
|
| 184 |
core_cot_score = summary["core_cot"]["macro_mean_score"]
|
| 185 |
row = {
|
| 186 |
-
"Models": get_display_model_name(model),
|
| 187 |
"Overall": round(summary["overall_score"] * 100, 2),
|
| 188 |
-
"Core
|
| 189 |
-
"Core
|
| 190 |
"Open-ended": round(summary["open"]["macro_mean_score"] * 100, 2)
|
| 191 |
}
|
| 192 |
for display_name in self.SUPER_GROUPS[selected_super_group]:
|
|
@@ -203,21 +231,54 @@ class DefaultDataLoader(BaseDataLoader):
|
|
| 203 |
|
| 204 |
def get_leaderboard_data(self, selected_super_group: str, selected_model_group: str) -> Tuple[list, list]:
|
| 205 |
df = self.get_df(selected_super_group, selected_model_group)
|
| 206 |
-
|
| 207 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 208 |
return headers, data
|
| 209 |
|
| 210 |
|
| 211 |
-
class
|
| 212 |
def __init__(self):
|
| 213 |
super().__init__()
|
| 214 |
|
| 215 |
def _load_model_data(self) -> Dict[str, Any]:
|
| 216 |
-
with open("./static/eval_results/
|
| 217 |
return json.load(f)
|
| 218 |
|
| 219 |
def _load_summary_data(self) -> Dict[str, Any]:
|
| 220 |
-
with open("./static/eval_results/
|
| 221 |
return json.load(f)
|
| 222 |
|
| 223 |
def get_df(self, selected_super_group: str, selected_model_group: str) -> pd.DataFrame:
|
|
@@ -226,10 +287,11 @@ class CoreSingleDataLoader(BaseDataLoader):
|
|
| 226 |
for model in self.MODEL_GROUPS[selected_model_group]:
|
| 227 |
model_data = self.MODEL_DATA[model]
|
| 228 |
summary = self.SUMMARY_DATA[model]
|
| 229 |
-
core_si_score = summary["macro_mean_score"]
|
| 230 |
row = {
|
| 231 |
-
"Models": get_display_model_name(model),
|
| 232 |
-
"
|
|
|
|
|
|
|
| 233 |
}
|
| 234 |
for display_name in self.SUPER_GROUPS[selected_super_group]:
|
| 235 |
original_keyword = self.keyword_display_map[display_name]
|
|
@@ -240,13 +302,43 @@ class CoreSingleDataLoader(BaseDataLoader):
|
|
| 240 |
data.append(row)
|
| 241 |
|
| 242 |
df = pd.DataFrame(data)
|
| 243 |
-
df = df.sort_values(by="
|
| 244 |
return df
|
| 245 |
|
| 246 |
def get_leaderboard_data(self, selected_super_group: str, selected_model_group: str) -> Tuple[list, list]:
|
| 247 |
df = self.get_df(selected_super_group, selected_model_group)
|
| 248 |
-
|
| 249 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 250 |
return headers, data
|
| 251 |
|
| 252 |
|
|
@@ -257,5 +349,8 @@ def get_original_dimension(mapped_dimension):
|
|
| 257 |
def get_original_keyword(mapped_keyword):
|
| 258 |
return next((k for k, v in KEYWORD_NAME_MAP.items() if v == mapped_keyword), mapped_keyword)
|
| 259 |
|
| 260 |
-
def get_display_model_name(model_name):
|
| 261 |
-
|
|
|
|
|
|
|
|
|
|
|
|
| 11 |
"InternVL2_76B": "InternVL2-Llama3-76B",
|
| 12 |
"Qwen2_VL_72B": "Qwen2-VL-72B",
|
| 13 |
"llava_onevision_72B": "Llava-OneVision-72B",
|
| 14 |
+
"NVLM": "NVLM-D-72B",
|
| 15 |
"GPT_4o_mini": "GPT-4o mini",
|
| 16 |
"Gemini_1.5_flash_002": "Gemini-1.5-Flash-002",
|
| 17 |
"Pixtral_12B": "Pixtral 12B",
|
|
|
|
| 83 |
"video": "Video",
|
| 84 |
}
|
| 85 |
|
| 86 |
+
MODEL_URLS = {
|
| 87 |
+
"Claude_3.5_new": "https://www.anthropic.com/news/3-5-models-and-computer-use",
|
| 88 |
+
"GPT_4o": "https://platform.openai.com/docs/models/gpt-4o",
|
| 89 |
+
"Claude_3.5": "https://www.anthropic.com/news/claude-3-5-sonnet",
|
| 90 |
+
"Gemini_1.5_pro_002": "https://ai.google.dev/gemini-api/docs/models/gemini",
|
| 91 |
+
"Gemini_1.5_flash_002": "https://ai.google.dev/gemini-api/docs/models/gemini",
|
| 92 |
+
"GPT_4o_mini": "https://platform.openai.com/docs/models#gpt-4o-mini",
|
| 93 |
+
"Qwen2_VL_72B": "https://huggingface.co/Qwen/Qwen2-VL-72B-Instruct",
|
| 94 |
+
"InternVL2_76B": "https://huggingface.co/OpenGVLab/InternVL2-Llama3-76B",
|
| 95 |
+
"llava_onevision_72B": "https://huggingface.co/lmms-lab/llava-onevision-qwen2-72b-ov-chat",
|
| 96 |
+
"NVLM": "https://huggingface.co/nvidia/NVLM-D-72B",
|
| 97 |
+
"Molmo_72B": "https://huggingface.co/allenai/Molmo-72B-0924",
|
| 98 |
+
"Qwen2_VL_7B": "https://huggingface.co/Qwen/Qwen2-VL-7B-Instruct",
|
| 99 |
+
"Pixtral_12B": "https://huggingface.co/mistralai/Pixtral-12B-2409",
|
| 100 |
+
"Aria": "https://huggingface.co/rhymes-ai/Aria",
|
| 101 |
+
"InternVL2_8B": "https://huggingface.co/OpenGVLab/InternVL2-8B",
|
| 102 |
+
"Phi-3.5-vision": "https://huggingface.co/microsoft/Phi-3.5-vision-instruct",
|
| 103 |
+
"MiniCPM_v2.6": "https://huggingface.co/openbmb/MiniCPM-V-2_6",
|
| 104 |
+
"llava_onevision_7B": "https://huggingface.co/lmms-lab/llava-onevision-qwen2-7b-ov",
|
| 105 |
+
"Llama_3_2_11B": "https://huggingface.co/meta-llama/Llama-3.2-11B-Vision",
|
| 106 |
+
"Idefics3": "https://huggingface.co/HuggingFaceM4/Idefics3-8B-Llama3",
|
| 107 |
+
"Molmo_7B_D": "https://huggingface.co/allenai/Molmo-7B-D-0924",
|
| 108 |
+
"Aquila_VL_2B": "https://huggingface.co/BAAI/Aquila-VL-2B-llava-qwen",
|
| 109 |
+
"POINTS_7B": "https://huggingface.co/WePOINTS/POINTS-Qwen-2-5-7B-Chat",
|
| 110 |
+
"Qwen2_VL_2B": "https://huggingface.co/Qwen/Qwen2-VL-2B-Instruct",
|
| 111 |
+
"InternVL2_2B": "https://huggingface.co/OpenGVLab/InternVL2-2B"
|
| 112 |
+
}
|
| 113 |
+
|
| 114 |
class BaseDataLoader:
|
| 115 |
# Define the base MODEL_GROUPS structure
|
| 116 |
BASE_MODEL_GROUPS = {
|
|
|
|
| 211 |
core_noncot_score = summary["core_noncot"]["macro_mean_score"]
|
| 212 |
core_cot_score = summary["core_cot"]["macro_mean_score"]
|
| 213 |
row = {
|
| 214 |
+
"Models": get_display_model_name(model, as_link=True),
|
| 215 |
"Overall": round(summary["overall_score"] * 100, 2),
|
| 216 |
+
"Core w/o CoT": round(core_noncot_score * 100, 2),
|
| 217 |
+
"Core w/ CoT": round(core_cot_score * 100, 2),
|
| 218 |
"Open-ended": round(summary["open"]["macro_mean_score"] * 100, 2)
|
| 219 |
}
|
| 220 |
for display_name in self.SUPER_GROUPS[selected_super_group]:
|
|
|
|
| 231 |
|
| 232 |
def get_leaderboard_data(self, selected_super_group: str, selected_model_group: str) -> Tuple[list, list]:
|
| 233 |
df = self.get_df(selected_super_group, selected_model_group)
|
| 234 |
+
|
| 235 |
+
# Get total task counts from the first model's data
|
| 236 |
+
sample_model = next(iter(self.MODEL_DATA))
|
| 237 |
+
total_core_tasks = self.SUMMARY_DATA[sample_model]["core_noncot"]["num_eval_tasks"]
|
| 238 |
+
total_open_tasks = self.SUMMARY_DATA[sample_model]["open"]["num_eval_tasks"]
|
| 239 |
+
total_tasks = total_core_tasks + total_open_tasks
|
| 240 |
+
|
| 241 |
+
# Define headers with task counts
|
| 242 |
+
column_headers = {
|
| 243 |
+
"Models": "Models",
|
| 244 |
+
"Overall": f"Overall ({total_tasks})",
|
| 245 |
+
"Core w/o CoT": f"Core(w/o CoT) ({total_core_tasks})",
|
| 246 |
+
"Core w/ CoT": f"Core(w/ CoT) ({total_core_tasks})",
|
| 247 |
+
"Open-ended": f"Open-ended ({total_open_tasks})"
|
| 248 |
+
}
|
| 249 |
+
|
| 250 |
+
# Rename the columns in DataFrame to match headers
|
| 251 |
+
df = df.rename(columns=column_headers)
|
| 252 |
+
|
| 253 |
+
headers = [
|
| 254 |
+
column_headers["Models"],
|
| 255 |
+
column_headers["Overall"],
|
| 256 |
+
column_headers["Core w/o CoT"],
|
| 257 |
+
column_headers["Core w/ CoT"],
|
| 258 |
+
column_headers["Open-ended"]
|
| 259 |
+
] + self.SUPER_GROUPS[selected_super_group]
|
| 260 |
+
|
| 261 |
+
data = df[[
|
| 262 |
+
column_headers["Models"],
|
| 263 |
+
column_headers["Overall"],
|
| 264 |
+
column_headers["Core w/o CoT"],
|
| 265 |
+
column_headers["Core w/ CoT"],
|
| 266 |
+
column_headers["Open-ended"]
|
| 267 |
+
] + self.SUPER_GROUPS[selected_super_group]].values.tolist()
|
| 268 |
+
|
| 269 |
return headers, data
|
| 270 |
|
| 271 |
|
| 272 |
+
class SingleImageDataLoader(BaseDataLoader):
|
| 273 |
def __init__(self):
|
| 274 |
super().__init__()
|
| 275 |
|
| 276 |
def _load_model_data(self) -> Dict[str, Any]:
|
| 277 |
+
with open("./static/eval_results/SI/all_model_keywords_stats.json", "r") as f:
|
| 278 |
return json.load(f)
|
| 279 |
|
| 280 |
def _load_summary_data(self) -> Dict[str, Any]:
|
| 281 |
+
with open("./static/eval_results/SI/all_summary.json", "r") as f:
|
| 282 |
return json.load(f)
|
| 283 |
|
| 284 |
def get_df(self, selected_super_group: str, selected_model_group: str) -> pd.DataFrame:
|
|
|
|
| 287 |
for model in self.MODEL_GROUPS[selected_model_group]:
|
| 288 |
model_data = self.MODEL_DATA[model]
|
| 289 |
summary = self.SUMMARY_DATA[model]
|
|
|
|
| 290 |
row = {
|
| 291 |
+
"Models": get_display_model_name(model, as_link=True),
|
| 292 |
+
"Overall": round(summary["overall_score"] * 100, 2),
|
| 293 |
+
"Core": round(summary["core"]["macro_mean_score"] * 100, 2),
|
| 294 |
+
"Open-ended": round(summary["open"]["macro_mean_score"] * 100, 2)
|
| 295 |
}
|
| 296 |
for display_name in self.SUPER_GROUPS[selected_super_group]:
|
| 297 |
original_keyword = self.keyword_display_map[display_name]
|
|
|
|
| 302 |
data.append(row)
|
| 303 |
|
| 304 |
df = pd.DataFrame(data)
|
| 305 |
+
df = df.sort_values(by="Overall", ascending=False)
|
| 306 |
return df
|
| 307 |
|
| 308 |
def get_leaderboard_data(self, selected_super_group: str, selected_model_group: str) -> Tuple[list, list]:
|
| 309 |
df = self.get_df(selected_super_group, selected_model_group)
|
| 310 |
+
|
| 311 |
+
# Get total task counts from the first model's data
|
| 312 |
+
sample_model = next(iter(self.MODEL_DATA))
|
| 313 |
+
total_core_tasks = self.SUMMARY_DATA[sample_model]["core"]["num_eval_tasks"]
|
| 314 |
+
total_open_tasks = self.SUMMARY_DATA[sample_model]["open"]["num_eval_tasks"]
|
| 315 |
+
total_tasks = total_core_tasks + total_open_tasks
|
| 316 |
+
|
| 317 |
+
# Define headers with task counts
|
| 318 |
+
column_headers = {
|
| 319 |
+
"Models": "Models",
|
| 320 |
+
"Overall": f"Overall ({total_tasks})",
|
| 321 |
+
"Core": f"Core ({total_core_tasks})",
|
| 322 |
+
"Open-ended": f"Open-ended ({total_open_tasks})"
|
| 323 |
+
}
|
| 324 |
+
|
| 325 |
+
# Rename the columns in DataFrame to match headers
|
| 326 |
+
df = df.rename(columns=column_headers)
|
| 327 |
+
|
| 328 |
+
headers = [
|
| 329 |
+
column_headers["Models"],
|
| 330 |
+
column_headers["Overall"],
|
| 331 |
+
column_headers["Core"],
|
| 332 |
+
column_headers["Open-ended"]
|
| 333 |
+
] + self.SUPER_GROUPS[selected_super_group]
|
| 334 |
+
|
| 335 |
+
data = df[[
|
| 336 |
+
column_headers["Models"],
|
| 337 |
+
column_headers["Overall"],
|
| 338 |
+
column_headers["Core"],
|
| 339 |
+
column_headers["Open-ended"]
|
| 340 |
+
] + self.SUPER_GROUPS[selected_super_group]].values.tolist()
|
| 341 |
+
|
| 342 |
return headers, data
|
| 343 |
|
| 344 |
|
|
|
|
| 349 |
def get_original_keyword(mapped_keyword):
|
| 350 |
return next((k for k, v in KEYWORD_NAME_MAP.items() if v == mapped_keyword), mapped_keyword)
|
| 351 |
|
| 352 |
+
def get_display_model_name(model_name: str, as_link: bool = True) -> str:
|
| 353 |
+
display_name = MODEL_NAME_MAP.get(model_name, model_name)
|
| 354 |
+
if as_link and model_name in MODEL_URLS:
|
| 355 |
+
return f'<a href="{MODEL_URLS[model_name]}" target="_blank" style="text-decoration: none; color: #2196F3;">{display_name}</a>'
|
| 356 |
+
return display_name
|