{"task":"image-sentence alignment","dataset":"VALSE counting adversarial","metric_names":["pairwise accuracy","Accuracy (%)"],"rows":[{"id":107532,"task":"image-sentence alignment","parent_task":"Multimodal Text and Image Classification","dataset":"VALSE counting adversarial","model_name":"ViLBERT 12-in-1","metrics":{"Accuracy (%)":"66.7","pairwise accuracy":"77.3"},"paper_url":"https://arxiv.org/abs/2112.07566v2","paper_title":"VALSE: A Task-Independent Benchmark for Vision and Language Models Centered on Linguistic Phenomena","paper_date":"2021-12-14","code_links":[{"title":"heidelberg-nlp/valse","url":"https://github.com/heidelberg-nlp/valse"}],"metrics_order":"[\"pairwise accuracy\", \"Accuracy (%)\"]","area":"Natural Language Processing","uses_additional_data":0,"source":"archive","tags":[]},{"id":107533,"task":"image-sentence alignment","parent_task":"Multimodal Text and Image Classification","dataset":"VALSE counting adversarial","model_name":"ViLBERT","metrics":{"Accuracy (%)":"51.8","pairwise accuracy":"73.7"},"paper_url":"https://arxiv.org/abs/2112.07566v2","paper_title":"VALSE: A Task-Independent Benchmark for Vision and Language Models Centered on Linguistic Phenomena","paper_date":"2021-12-14","code_links":[{"title":"heidelberg-nlp/valse","url":"https://github.com/heidelberg-nlp/valse"}],"metrics_order":"[\"pairwise accuracy\", \"Accuracy (%)\"]","area":"Natural Language Processing","uses_additional_data":0,"source":"archive","tags":[]},{"id":107534,"task":"image-sentence alignment","parent_task":"Multimodal Text and Image Classification","dataset":"VALSE counting adversarial","model_name":"GPT1","metrics":{"pairwise accuracy":"69.5"},"paper_url":"https://arxiv.org/abs/2112.07566v2","paper_title":"VALSE: A Task-Independent Benchmark for Vision and Language Models Centered on Linguistic Phenomena","paper_date":"2021-12-14","code_links":[{"title":"heidelberg-nlp/valse","url":"https://github.com/heidelberg-nlp/valse"}],"metrics_order":"[\"pairwise accuracy\", \"Accuracy (%)\"]","area":"Natural Language Processing","uses_additional_data":0,"source":"archive","tags":[]},{"id":107535,"task":"image-sentence alignment","parent_task":"Multimodal Text and Image Classification","dataset":"VALSE counting adversarial","model_name":"CLIP","metrics":{"pairwise accuracy":"57.5"},"paper_url":"https://arxiv.org/abs/2112.07566v2","paper_title":"VALSE: A Task-Independent Benchmark for Vision and Language Models Centered on Linguistic Phenomena","paper_date":"2021-12-14","code_links":[{"title":"heidelberg-nlp/valse","url":"https://github.com/heidelberg-nlp/valse"}],"metrics_order":"[\"pairwise accuracy\", \"Accuracy (%)\"]","area":"Natural Language Processing","uses_additional_data":0,"source":"archive","tags":[]},{"id":107536,"task":"image-sentence alignment","parent_task":"Multimodal Text and Image Classification","dataset":"VALSE counting adversarial","model_name":"VisualBERT","metrics":{"Accuracy (%)":"50.0","pairwise accuracy":"50.0"},"paper_url":"https://arxiv.org/abs/2112.07566v2","paper_title":"VALSE: A Task-Independent Benchmark for Vision and Language Models Centered on Linguistic Phenomena","paper_date":"2021-12-14","code_links":[{"title":"heidelberg-nlp/valse","url":"https://github.com/heidelberg-nlp/valse"}],"metrics_order":"[\"pairwise accuracy\", \"Accuracy (%)\"]","area":"Natural Language Processing","uses_additional_data":0,"source":"archive","tags":[]},{"id":107537,"task":"image-sentence alignment","parent_task":"Multimodal Text and Image Classification","dataset":"VALSE counting adversarial","model_name":"GPT2","metrics":{"pairwise accuracy":"45.3"},"paper_url":"https://arxiv.org/abs/2112.07566v2","paper_title":"VALSE: A Task-Independent Benchmark for Vision and Language Models Centered on Linguistic Phenomena","paper_date":"2021-12-14","code_links":[{"title":"heidelberg-nlp/valse","url":"https://github.com/heidelberg-nlp/valse"}],"metrics_order":"[\"pairwise accuracy\", \"Accuracy (%)\"]","area":"Natural Language Processing","uses_additional_data":0,"source":"archive","tags":[]},{"id":107538,"task":"image-sentence alignment","parent_task":"Multimodal Text and Image Classification","dataset":"VALSE counting adversarial","model_name":"LXMERT","metrics":{"Accuracy (%)":"49.9","pairwise accuracy":"42.6"},"paper_url":"https://arxiv.org/abs/2112.07566v2","paper_title":"VALSE: A Task-Independent Benchmark for Vision and Language Models Centered on Linguistic Phenomena","paper_date":"2021-12-14","code_links":[{"title":"heidelberg-nlp/valse","url":"https://github.com/heidelberg-nlp/valse"}],"metrics_order":"[\"pairwise accuracy\", \"Accuracy (%)\"]","area":"Natural Language Processing","uses_additional_data":0,"source":"archive","tags":[]},{"id":129456,"task":"image-sentence alignment","parent_task":"Multimodal Text and Image Classification","dataset":"VALSE counting adversarial","model_name":"ViLBERT 12-in-1","metrics":{"Accuracy (%)":"66.7","pairwise accuracy":"77.3"},"paper_url":"https://arxiv.org/abs/2112.07566v2","paper_title":"VALSE: A Task-Independent Benchmark for Vision and Language Models Centered on Linguistic Phenomena","paper_date":"2021-12-14","code_links":[{"title":"heidelberg-nlp/valse","url":"https://github.com/heidelberg-nlp/valse"}],"metrics_order":"[\"pairwise accuracy\", \"Accuracy (%)\"]","area":"Natural Language Processing","uses_additional_data":0,"source":"archive","tags":[]},{"id":129457,"task":"image-sentence alignment","parent_task":"Multimodal Text and Image Classification","dataset":"VALSE counting adversarial","model_name":"ViLBERT","metrics":{"Accuracy (%)":"51.8","pairwise accuracy":"73.7"},"paper_url":"https://arxiv.org/abs/2112.07566v2","paper_title":"VALSE: A Task-Independent Benchmark for Vision and Language Models Centered on Linguistic Phenomena","paper_date":"2021-12-14","code_links":[{"title":"heidelberg-nlp/valse","url":"https://github.com/heidelberg-nlp/valse"}],"metrics_order":"[\"pairwise accuracy\", \"Accuracy (%)\"]","area":"Natural Language Processing","uses_additional_data":0,"source":"archive","tags":[]},{"id":129458,"task":"image-sentence alignment","parent_task":"Multimodal Text and Image Classification","dataset":"VALSE counting adversarial","model_name":"GPT1","metrics":{"pairwise accuracy":"69.5"},"paper_url":"https://arxiv.org/abs/2112.07566v2","paper_title":"VALSE: A Task-Independent Benchmark for Vision and Language Models Centered on Linguistic Phenomena","paper_date":"2021-12-14","code_links":[{"title":"heidelberg-nlp/valse","url":"https://github.com/heidelberg-nlp/valse"}],"metrics_order":"[\"pairwise accuracy\", \"Accuracy (%)\"]","area":"Natural Language Processing","uses_additional_data":0,"source":"archive","tags":[]},{"id":129459,"task":"image-sentence alignment","parent_task":"Multimodal Text and Image Classification","dataset":"VALSE counting adversarial","model_name":"CLIP","metrics":{"pairwise accuracy":"57.5"},"paper_url":"https://arxiv.org/abs/2112.07566v2","paper_title":"VALSE: A Task-Independent Benchmark for Vision and Language Models Centered on Linguistic Phenomena","paper_date":"2021-12-14","code_links":[{"title":"heidelberg-nlp/valse","url":"https://github.com/heidelberg-nlp/valse"}],"metrics_order":"[\"pairwise accuracy\", \"Accuracy (%)\"]","area":"Natural Language Processing","uses_additional_data":0,"source":"archive","tags":[]},{"id":129460,"task":"image-sentence alignment","parent_task":"Multimodal Text and Image Classification","dataset":"VALSE counting adversarial","model_name":"VisualBERT","metrics":{"Accuracy (%)":"50.0","pairwise accuracy":"50.0"},"paper_url":"https://arxiv.org/abs/2112.07566v2","paper_title":"VALSE: A Task-Independent Benchmark for Vision and Language Models Centered on Linguistic Phenomena","paper_date":"2021-12-14","code_links":[{"title":"heidelberg-nlp/valse","url":"https://github.com/heidelberg-nlp/valse"}],"metrics_order":"[\"pairwise accuracy\", \"Accuracy (%)\"]","area":"Natural Language Processing","uses_additional_data":0,"source":"archive","tags":[]},{"id":129461,"task":"image-sentence alignment","parent_task":"Multimodal Text and Image Classification","dataset":"VALSE counting adversarial","model_name":"GPT2","metrics":{"pairwise accuracy":"45.3"},"paper_url":"https://arxiv.org/abs/2112.07566v2","paper_title":"VALSE: A Task-Independent Benchmark for Vision and Language Models Centered on Linguistic Phenomena","paper_date":"2021-12-14","code_links":[{"title":"heidelberg-nlp/valse","url":"https://github.com/heidelberg-nlp/valse"}],"metrics_order":"[\"pairwise accuracy\", \"Accuracy (%)\"]","area":"Natural Language Processing","uses_additional_data":0,"source":"archive","tags":[]},{"id":129462,"task":"image-sentence alignment","parent_task":"Multimodal Text and Image Classification","dataset":"VALSE counting adversarial","model_name":"LXMERT","metrics":{"Accuracy (%)":"49.9","pairwise accuracy":"42.6"},"paper_url":"https://arxiv.org/abs/2112.07566v2","paper_title":"VALSE: A Task-Independent Benchmark for Vision and Language Models Centered on Linguistic Phenomena","paper_date":"2021-12-14","code_links":[{"title":"heidelberg-nlp/valse","url":"https://github.com/heidelberg-nlp/valse"}],"metrics_order":"[\"pairwise accuracy\", \"Accuracy (%)\"]","area":"Natural Language Processing","uses_additional_data":0,"source":"archive","tags":[]}]}