{"task":"image-sentence alignment","dataset":"VALSE spatial relations","metric_names":["pairwise accuracy","Accuracy (%)"],"rows":[{"id":107581,"task":"image-sentence alignment","parent_task":"Multimodal Text and Image Classification","dataset":"VALSE spatial relations","model_name":"GPT1","metrics":{"pairwise accuracy":"77.2"},"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":107582,"task":"image-sentence alignment","parent_task":"Multimodal Text and Image Classification","dataset":"VALSE spatial relations","model_name":"GPT2","metrics":{"pairwise accuracy":"75.0"},"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":107583,"task":"image-sentence alignment","parent_task":"Multimodal Text and Image Classification","dataset":"VALSE spatial relations","model_name":"ViLBERT 12-in-1","metrics":{"Accuracy (%)":"53.4","pairwise accuracy":"67.7"},"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":107584,"task":"image-sentence alignment","parent_task":"Multimodal Text and Image Classification","dataset":"VALSE spatial relations","model_name":"CLIP","metrics":{"pairwise accuracy":"64.3"},"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":107585,"task":"image-sentence alignment","parent_task":"Multimodal Text and Image Classification","dataset":"VALSE spatial relations","model_name":"LXMERT","metrics":{"Accuracy (%)":"50.8","pairwise accuracy":"60.2"},"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":107586,"task":"image-sentence alignment","parent_task":"Multimodal Text and Image Classification","dataset":"VALSE spatial relations","model_name":"ViLBERT","metrics":{"Accuracy (%)":"49.9","pairwise accuracy":"57.2"},"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":107587,"task":"image-sentence alignment","parent_task":"Multimodal Text and Image Classification","dataset":"VALSE spatial relations","model_name":"VisualBERT","metrics":{"Accuracy (%)":"49.3","pairwise accuracy":"39.7"},"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":129505,"task":"image-sentence alignment","parent_task":"Multimodal Text and Image Classification","dataset":"VALSE spatial relations","model_name":"GPT1","metrics":{"pairwise accuracy":"77.2"},"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":129506,"task":"image-sentence alignment","parent_task":"Multimodal Text and Image Classification","dataset":"VALSE spatial relations","model_name":"GPT2","metrics":{"pairwise accuracy":"75.0"},"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":129507,"task":"image-sentence alignment","parent_task":"Multimodal Text and Image Classification","dataset":"VALSE spatial relations","model_name":"ViLBERT 12-in-1","metrics":{"Accuracy (%)":"53.4","pairwise accuracy":"67.7"},"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":129508,"task":"image-sentence alignment","parent_task":"Multimodal Text and Image Classification","dataset":"VALSE spatial relations","model_name":"CLIP","metrics":{"pairwise accuracy":"64.3"},"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":129509,"task":"image-sentence alignment","parent_task":"Multimodal Text and Image Classification","dataset":"VALSE spatial relations","model_name":"LXMERT","metrics":{"Accuracy (%)":"50.8","pairwise accuracy":"60.2"},"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":129510,"task":"image-sentence alignment","parent_task":"Multimodal Text and Image Classification","dataset":"VALSE spatial relations","model_name":"ViLBERT","metrics":{"Accuracy (%)":"49.9","pairwise accuracy":"57.2"},"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":129511,"task":"image-sentence alignment","parent_task":"Multimodal Text and Image Classification","dataset":"VALSE spatial relations","model_name":"VisualBERT","metrics":{"Accuracy (%)":"49.3","pairwise accuracy":"39.7"},"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":[]}]}