Image Sentence Alignment On Valse
Metrics
Average Accuracy
average pairwise accuracy
Results
Performance results of various models on this benchmark
Model Name | Average Accuracy | average pairwise accuracy | Paper Title | Repository |
|---|---|---|---|---|
| ViLBERT 12-in-1 | 63.2 | 75.1 | VALSE: A Task-Independent Benchmark for Vision and Language Models Centered on Linguistic Phenomena | - |
| LXMERT | 53.5 | 59.6 | VALSE: A Task-Independent Benchmark for Vision and Language Models Centered on Linguistic Phenomena | - |
| CLIP | - | 64.0 | VALSE: A Task-Independent Benchmark for Vision and Language Models Centered on Linguistic Phenomena | - |
| ViLBERT | 51.3 | 63.7 | VALSE: A Task-Independent Benchmark for Vision and Language Models Centered on Linguistic Phenomena | - |
| VisualBERT | 48.8 | 46.4 | VALSE: A Task-Independent Benchmark for Vision and Language Models Centered on Linguistic Phenomena | - |
| GPT1 | - | 60.7 | VALSE: A Task-Independent Benchmark for Vision and Language Models Centered on Linguistic Phenomena | - |
| GPT2 | - | 60.1 | VALSE: A Task-Independent Benchmark for Vision and Language Models Centered on Linguistic Phenomena | - |
0 of 7 row(s) selected.