{"task":"image-sentence alignment","dataset":"VALSE coreference standard","metric_names":["pairwise accuracy","Accuracy (%)"],"rows":[{"id":107574,"task":"image-sentence alignment","parent_task":"Multimodal Text and Image Classification","dataset":"VALSE coreference standard","model_name":"ViLBERT 12-in-1","metrics":{"Accuracy (%)":"54.4","pairwise accuracy":"75.7"},"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":107575,"task":"image-sentence alignment","parent_task":"Multimodal Text and Image Classification","dataset":"VALSE coreference standard","model_name":"GPT2","metrics":{"pairwise accuracy":"54.5"},"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":107576,"task":"image-sentence alignment","parent_task":"Multimodal Text and Image Classification","dataset":"VALSE coreference standard","model_name":"CLIP","metrics":{"pairwise accuracy":"52.1"},"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":107577,"task":"image-sentence alignment","parent_task":"Multimodal Text and Image Classification","dataset":"VALSE coreference standard","model_name":"VisualBERT","metrics":{"Accuracy (%)":"50.0","pairwise accuracy":"49.5"},"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":107578,"task":"image-sentence alignment","parent_task":"Multimodal Text and Image Classification","dataset":"VALSE coreference standard","model_name":"ViLBERT","metrics":{"Accuracy (%)":"50.0","pairwise accuracy":"47.2"},"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":107579,"task":"image-sentence alignment","parent_task":"Multimodal Text and Image Classification","dataset":"VALSE coreference standard","model_name":"LXMERT","metrics":{"Accuracy (%)":"49.8","pairwise accuracy":"46.8"},"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":107580,"task":"image-sentence alignment","parent_task":"Multimodal Text and Image Classification","dataset":"VALSE coreference standard","model_name":"GPT1","metrics":{"pairwise accuracy":"45.6"},"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":129498,"task":"image-sentence alignment","parent_task":"Multimodal Text and Image Classification","dataset":"VALSE coreference standard","model_name":"ViLBERT 12-in-1","metrics":{"Accuracy (%)":"54.4","pairwise accuracy":"75.7"},"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":129499,"task":"image-sentence alignment","parent_task":"Multimodal Text and Image Classification","dataset":"VALSE coreference standard","model_name":"GPT2","metrics":{"pairwise accuracy":"54.5"},"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":129500,"task":"image-sentence alignment","parent_task":"Multimodal Text and Image Classification","dataset":"VALSE coreference standard","model_name":"CLIP","metrics":{"pairwise accuracy":"52.1"},"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":129501,"task":"image-sentence alignment","parent_task":"Multimodal Text and Image Classification","dataset":"VALSE coreference standard","model_name":"VisualBERT","metrics":{"Accuracy (%)":"50.0","pairwise accuracy":"49.5"},"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":129502,"task":"image-sentence alignment","parent_task":"Multimodal Text and Image Classification","dataset":"VALSE coreference standard","model_name":"ViLBERT","metrics":{"Accuracy (%)":"50.0","pairwise accuracy":"47.2"},"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":129503,"task":"image-sentence alignment","parent_task":"Multimodal Text and Image Classification","dataset":"VALSE coreference standard","model_name":"LXMERT","metrics":{"Accuracy (%)":"49.8","pairwise accuracy":"46.8"},"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":129504,"task":"image-sentence alignment","parent_task":"Multimodal Text and Image Classification","dataset":"VALSE coreference standard","model_name":"GPT1","metrics":{"pairwise accuracy":"45.6"},"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":[]}]}