image-sentence alignment on VALSE counting small numbers
pairwise accuracy
- 2021-12-14 — ViLBERT 12-in-1: pairwise accuracy 80.2
| Rank | Model | pairwise accuracy | Accuracy (%) | Paper | Code | Year |
|---|---|---|---|---|---|---|
| 11 | CLIP | 62.5 | – | VALSE: A Task-Independent Benchmark for Vision and Language Models Centered on Linguistic Phenomena | heidelberg-nlp/valse | 2021 |
| 12 | GPT2 | 49.8 | – | VALSE: A Task-Independent Benchmark for Vision and Language Models Centered on Linguistic Phenomena | heidelberg-nlp/valse | 2021 |
| 13 | GPT1 | 48.7 | – | VALSE: A Task-Independent Benchmark for Vision and Language Models Centered on Linguistic Phenomena | heidelberg-nlp/valse | 2021 |
| 14 | VisualBERT | 48.2 | 47.8 | VALSE: A Task-Independent Benchmark for Vision and Language Models Centered on Linguistic Phenomena | heidelberg-nlp/valse | 2021 |