Datasets:
ibm
/

Modalities:
Text
Formats:
csv
Languages:
English
Size:
< 1K
ArXiv:
Libraries:
Datasets
pandas
License:
Files changed (1) hide show
  1. README.md +19 -1
README.md CHANGED
@@ -63,6 +63,24 @@ Each instance consists of a question, a pair of contradictory passages extracted
63
 
64
  The dataset has been used in the paper to assess LLMs performance when augmented with retrieved passages containing real-world knowledge conflicts.
65
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
66
  ### Out-of-Scope Use
67
 
68
  <!-- This section addresses misuse, malicious use, and uses that the dataset will not work well for. -->
@@ -203,7 +221,7 @@ For further information, see the annotation guidelines of the paper.
203
 
204
  <!-- This section describes the people or systems who created the annotations. -->
205
 
206
- Yufang Hou, Alessandra Pascale, Javier Carnerero-Cano, Tigran Tchrakian, Radu Marinescu, Elizabeth Daly, Inkit Padhi, and Prasanna Sattigeri.
207
 
208
  #### Personal and Sensitive Information
209
 
 
63
 
64
  The dataset has been used in the paper to assess LLMs performance when augmented with retrieved passages containing real-world knowledge conflicts.
65
 
66
+ The following figure illustrates the evaluation process:
67
+
68
+ <p align="center">
69
+ <img src="./figs/Evaluation.png" width=70%/>
70
+ </p>
71
+
72
+ And the following table shows the performance of five LLMs (Mistral-7b-inst, Mixtral-8x7b-inst, Llama-2-70b-chat, Llama-3-70b-inst, and GPT-4) on the Wikipedia Contradict Benchmark based on rigorous human evaluations on a subset of answers for 55 instances, which corresponds to 1,375 LLM responses in total.
73
+
74
+ <p align="center">
75
+ <img src="./figs/table2.png" width=70%/>
76
+ </p>
77
+
78
+ Notes: “C”, “PC” and “IC” stand for “Correct”, “Partially correct”, “Incorrect”, respectively. “all”, “exp”, and “imp” represent for instance
79
+ types: all instances, instances with explicit conflicts, and instances with implicit conflicts. The
80
+ numbers represent the ratio of responses from each LLM that were assessed as “Correct, “Partially
81
+ correct, or “Incorrect for each instance type under a prompt template. The bold numbers highlight
82
+ the best models that correctly answer questions for each type and prompt template.
83
+
84
  ### Out-of-Scope Use
85
 
86
  <!-- This section addresses misuse, malicious use, and uses that the dataset will not work well for. -->
 
221
 
222
  <!-- This section describes the people or systems who created the annotations. -->
223
 
224
+ Yufang Hou, Alessandra Pascale, Javier Carnerero-Cano, Tigran Tchrakian, Radu Marinescu, Elizabeth Daly, Inkit Padhi
225
 
226
  #### Personal and Sensitive Information
227