fix legend + display
Browse files
README.md
CHANGED
|
@@ -243,15 +243,15 @@ As opposed to Flamingo, we did not train IDEFICS on video-text pairs datasets, a
|
|
| 243 |
|
| 244 |
We note that since IDEFICS was trained on PMD (which contains COCO), the evaluation numbers on COCO are not directly comparable with Flamingo and OpenFlamingo since they did not explicitely have this dataset in the training mixture. Additionally, Flamingo is trained with images of resolution 320 x 320 while IDEFICS and OpenFlamingo were trained with images of 224 x 224 resolution.
|
| 245 |
|
| 246 |
-
| Model | Shots | <nobr>VQAv2<br>OE VQA acc.</nobr> | <nobr>OKVQA<br>OE VQA acc.</nobr> | <nobr>TextVQA<br>OE VQA acc.</nobr> | <nobr>VizWiz<br>OE VQA acc.</nobr> | <nobr>TextCaps<br>CIDEr</nobr> | <nobr>Coco<br>CIDEr</nobr> | <nobr>NoCaps<br>CIDEr</nobr> | <nobr>Flickr<br>CIDEr</nobr> | <nobr>VisDial<br>NDCG</nobr> | <nobr>HatefulMemes<br>ROC AUC</nobr> | <nobr>ScienceQA<br>acc.</nobr> | <nobr>RenderedSST2<br>acc.</nobr> | <nobr>Winoground<br>group
|
| 247 |
|:------------|--------:|---------------------:|---------------------:|-----------------------:|----------------------:|-------------------:|---------------:|-----------------:|-----------------:|-----------------:|-------------------------:|-----------------------:|--------------------------:|----------------------------------:|
|
| 248 |
-
| IDEFICS 80B | 0 | 60.0 | 45.2 | 30.9 | 36.0 | 56.8 | 91.8 | 65.0 | 53.7 | 48.8 | 60.6 | 68.9 | 60.5 | 8.0
|
| 249 |
| | 4 | 63.6 | 52.4 | 34.4 | 40.4 | 72.7 | 110.3 | 99.6 | 73.7 | 48.4 | 57.8 | 58.9 | 66.6 | - |
|
| 250 |
| | 8 | 64.8 | 55.1 | 35.7 | 46.1 | 77.6 | 114.3 | 105.7 | 76.6 | 47.9 | 58.2 | - | 67.8 | - |
|
| 251 |
| | 16 | 65.4 | 56.8 | 36.3 | 48.3 | 81.4 | 116.6 | 107.0 | 80.1 | - | 55.8 | - | 67.7 | - |
|
| 252 |
| | 32 | 65.9 | 57.8 | 36.7 | 50.0 | 82.7 | 116.6 | 107.5 | 81.1 | - | 52.5 | - | 67.3 | - |
|
| 253 |
<br>
|
| 254 |
-
| IDEFICS 9B | 0 | 50.9 | 38.4 | 25.9 | 35.5 | 25.4 | 46.0 | 36.8 | 27.3 | 48.7 | 51.7 | 44.2 | 61.8 | 5.0
|
| 255 |
| | 4 | 55.4 | 45.5 | 27.6 | 36.9 | 60.0 | 93.0 | 81.3 | 59.7 | 47.9 | 50.7 | 37.4 | 62.3 | - |
|
| 256 |
| | 8 | 56.4 | 47.7 | 27.5 | 40.4 | 63.2 | 97.0 | 86.8 | 61.9 | 47.6 | 51.0 | - | 66.3 | - |
|
| 257 |
| | 16 | 57.0 | 48.4 | 27.9 | 42.6 | 67.4 | 99.7 | 89.4 | 64.5 | - | 50.9 | - | 67.8 | - |
|
|
@@ -271,7 +271,7 @@ For ImageNet-1k, we also report results where the priming samples are selected t
|
|
| 271 |
|
| 272 |
Similarly to the base IDEFICS models, we performed checkpoint selection to stop the training. Given that M3IT contains in the training set a handful of the benchmarks we were evaluating on, we used [MMBench](https://huggingface.co/papers/2307.06281) as a held-out validation benchmark to perform checkpoint selection. We select the checkpoint at step 3'000 for IDEFICS-80b-instruct and at step 8'000 for IDEFICS-9b-instruct.
|
| 273 |
|
| 274 |
-
| Model | Shots | <nobr>VQAv2 <br>OE VQA acc.</nobr> | <nobr>OKVQA <br>OE VQA acc.</nobr> | <nobr>TextVQA <br>OE VQA acc.</nobr> | <nobr>VizWiz<br>OE VQA acc.</nobr> | <nobr>TextCaps <br>CIDEr</nobr> | <nobr>Coco <br>CIDEr</nobr> | <nobr>NoCaps<br>CIDEr</nobr> | <nobr>Flickr<br>CIDEr</nobr> | <nobr>VisDial <br>NDCG</nobr> | <nobr>HatefulMemes<br>ROC AUC</nobr> | <nobr>ScienceQA <br>acc.</nobr> | <nobr>RenderedSST2<br>acc.</nobr> | <nobr>Winoground<br>group
|
| 275 |
| :--------------------- | --------: | ---------------------: | ---------------------: | -----------------------: | ----------------------: | -------------------: | ---------------: | -----------------: | -----------------: | -----------------: | -------------------------: | -----------------------: | --------------------------: | ----------------------------------: |
|
| 276 |
| Finetuning data **does not** contain the evaluation dataset | - | ✖ | ✖ | ✖ | ✔ | ✖ | ✖ | ✖ | ✔ | ✖ | ✔ | ✖ | ✔ | ✖ |
|
| 277 |
| <nobr>IDEFICS 80B Instruct<br> | 0 | 37.4 (-22.7) | 36.9 (-8.2) | 32.9 (1.9) | 26.2 (-9.8) | 76.5 (19.7) | 117.2 (25.4) | 104.5 (39.5) | 65.3 (11.7) | 49.3 (0.4) | 58.9 (-1.7) | 69.5 (0.5) | 67.3 (6.8) | 9.2/20.0/25.0 (1.2/1.2/2.5) |
|
|
@@ -286,6 +286,7 @@ Similarly to the base IDEFICS models, we performed checkpoint selection to stop
|
|
| 286 |
| | 16 | 66.8 (9.8) | 51.7 (3.3) | 31.6 (3.7) | 44.8 (2.3) | 70.2 (2.7) | 128.8 (29.1) | 101.5 (12.2) | 75.8 (11.4) | - | 51.7 (0.7) | - | 63.3 (-4.6) | - |
|
| 287 |
| | 32 | 66.9 (9.0) | 52.3 (2.7) | 32.0 (3.7) | 46.0 (2.2) | 71.7 (3.6) | 127.8 (29.8) | 101.0 (10.5) | 76.3 (11.9) | - | 50.8 (1.0) | - | 60.9 (-6.1) | - |
|
| 288 |
|
|
|
|
| 289 |
|
| 290 |
# Technical Specifications
|
| 291 |
|
|
|
|
| 243 |
|
| 244 |
We note that since IDEFICS was trained on PMD (which contains COCO), the evaluation numbers on COCO are not directly comparable with Flamingo and OpenFlamingo since they did not explicitely have this dataset in the training mixture. Additionally, Flamingo is trained with images of resolution 320 x 320 while IDEFICS and OpenFlamingo were trained with images of 224 x 224 resolution.
|
| 245 |
|
| 246 |
+
| Model | Shots | <nobr>VQAv2<br>OE VQA acc.</nobr> | <nobr>OKVQA<br>OE VQA acc.</nobr> | <nobr>TextVQA<br>OE VQA acc.</nobr> | <nobr>VizWiz<br>OE VQA acc.</nobr> | <nobr>TextCaps<br>CIDEr</nobr> | <nobr>Coco<br>CIDEr</nobr> | <nobr>NoCaps<br>CIDEr</nobr> | <nobr>Flickr<br>CIDEr</nobr> | <nobr>VisDial<br>NDCG</nobr> | <nobr>HatefulMemes<br>ROC AUC</nobr> | <nobr>ScienceQA<br>acc.</nobr> | <nobr>RenderedSST2<br>acc.</nobr> | <nobr>Winoground<br>group/text/image</nobr> |
|
| 247 |
|:------------|--------:|---------------------:|---------------------:|-----------------------:|----------------------:|-------------------:|---------------:|-----------------:|-----------------:|-----------------:|-------------------------:|-----------------------:|--------------------------:|----------------------------------:|
|
| 248 |
+
| IDEFICS 80B | 0 | 60.0 | 45.2 | 30.9 | 36.0 | 56.8 | 91.8 | 65.0 | 53.7 | 48.8 | 60.6 | 68.9 | 60.5 | 8.0/18.75/22.5|
|
| 249 |
| | 4 | 63.6 | 52.4 | 34.4 | 40.4 | 72.7 | 110.3 | 99.6 | 73.7 | 48.4 | 57.8 | 58.9 | 66.6 | - |
|
| 250 |
| | 8 | 64.8 | 55.1 | 35.7 | 46.1 | 77.6 | 114.3 | 105.7 | 76.6 | 47.9 | 58.2 | - | 67.8 | - |
|
| 251 |
| | 16 | 65.4 | 56.8 | 36.3 | 48.3 | 81.4 | 116.6 | 107.0 | 80.1 | - | 55.8 | - | 67.7 | - |
|
| 252 |
| | 32 | 65.9 | 57.8 | 36.7 | 50.0 | 82.7 | 116.6 | 107.5 | 81.1 | - | 52.5 | - | 67.3 | - |
|
| 253 |
<br>
|
| 254 |
+
| IDEFICS 9B | 0 | 50.9 | 38.4 | 25.9 | 35.5 | 25.4 | 46.0 | 36.8 | 27.3 | 48.7 | 51.7 | 44.2 | 61.8 | 5.0 /16.8/20.8 |
|
| 255 |
| | 4 | 55.4 | 45.5 | 27.6 | 36.9 | 60.0 | 93.0 | 81.3 | 59.7 | 47.9 | 50.7 | 37.4 | 62.3 | - |
|
| 256 |
| | 8 | 56.4 | 47.7 | 27.5 | 40.4 | 63.2 | 97.0 | 86.8 | 61.9 | 47.6 | 51.0 | - | 66.3 | - |
|
| 257 |
| | 16 | 57.0 | 48.4 | 27.9 | 42.6 | 67.4 | 99.7 | 89.4 | 64.5 | - | 50.9 | - | 67.8 | - |
|
|
|
|
| 271 |
|
| 272 |
Similarly to the base IDEFICS models, we performed checkpoint selection to stop the training. Given that M3IT contains in the training set a handful of the benchmarks we were evaluating on, we used [MMBench](https://huggingface.co/papers/2307.06281) as a held-out validation benchmark to perform checkpoint selection. We select the checkpoint at step 3'000 for IDEFICS-80b-instruct and at step 8'000 for IDEFICS-9b-instruct.
|
| 273 |
|
| 274 |
+
| Model | Shots | <nobr>VQAv2 <br>OE VQA acc.</nobr> | <nobr>OKVQA <br>OE VQA acc.</nobr> | <nobr>TextVQA <br>OE VQA acc.</nobr> | <nobr>VizWiz<br>OE VQA acc.</nobr> | <nobr>TextCaps <br>CIDEr</nobr> | <nobr>Coco <br>CIDEr</nobr> | <nobr>NoCaps<br>CIDEr</nobr> | <nobr>Flickr<br>CIDEr</nobr> | <nobr>VisDial <br>NDCG</nobr> | <nobr>HatefulMemes<br>ROC AUC</nobr> | <nobr>ScienceQA <br>acc.</nobr> | <nobr>RenderedSST2<br>acc.</nobr> | <nobr>Winoground<br>group/text/image</nobr> |
|
| 275 |
| :--------------------- | --------: | ---------------------: | ---------------------: | -----------------------: | ----------------------: | -------------------: | ---------------: | -----------------: | -----------------: | -----------------: | -------------------------: | -----------------------: | --------------------------: | ----------------------------------: |
|
| 276 |
| Finetuning data **does not** contain the evaluation dataset | - | ✖ | ✖ | ✖ | ✔ | ✖ | ✖ | ✖ | ✔ | ✖ | ✔ | ✖ | ✔ | ✖ |
|
| 277 |
| <nobr>IDEFICS 80B Instruct<br> | 0 | 37.4 (-22.7) | 36.9 (-8.2) | 32.9 (1.9) | 26.2 (-9.8) | 76.5 (19.7) | 117.2 (25.4) | 104.5 (39.5) | 65.3 (11.7) | 49.3 (0.4) | 58.9 (-1.7) | 69.5 (0.5) | 67.3 (6.8) | 9.2/20.0/25.0 (1.2/1.2/2.5) |
|
|
|
|
| 286 |
| | 16 | 66.8 (9.8) | 51.7 (3.3) | 31.6 (3.7) | 44.8 (2.3) | 70.2 (2.7) | 128.8 (29.1) | 101.5 (12.2) | 75.8 (11.4) | - | 51.7 (0.7) | - | 63.3 (-4.6) | - |
|
| 287 |
| | 32 | 66.9 (9.0) | 52.3 (2.7) | 32.0 (3.7) | 46.0 (2.2) | 71.7 (3.6) | 127.8 (29.8) | 101.0 (10.5) | 76.3 (11.9) | - | 50.8 (1.0) | - | 60.9 (-6.1) | - |
|
| 288 |
|
| 289 |
+
*() Improvement over non-instruct version.
|
| 290 |
|
| 291 |
# Technical Specifications
|
| 292 |
|