{"task":"Common Sense Reasoning","dataset":"PARus","metric_names":["Accuracy"],"rows":[{"id":32196,"task":"Common Sense Reasoning","parent_task":null,"dataset":"PARus","model_name":"Human Benchmark","metrics":{"Accuracy":"0.982"},"paper_url":"https://arxiv.org/abs/2010.15925v2","paper_title":"RussianSuperGLUE: A Russian Language Understanding Evaluation Benchmark","paper_date":"2020-10-29","code_links":[{"title":"RussianNLP/RussianSuperGLUE","url":"https://github.com/RussianNLP/RussianSuperGLUE"},{"title":"RussianNLP/MOROCCO","url":"https://github.com/RussianNLP/MOROCCO"}],"metrics_order":"[\"Accuracy\"]","area":"Natural Language Processing","uses_additional_data":0,"source":"archive","tags":[]},{"id":32197,"task":"Common Sense Reasoning","parent_task":null,"dataset":"PARus","model_name":"Golden Transformer","metrics":{"Accuracy":"0.908"},"paper_url":null,"paper_title":null,"paper_date":null,"code_links":[],"metrics_order":"[\"Accuracy\"]","area":"Natural Language Processing","uses_additional_data":0,"source":"archive","tags":[]},{"id":32198,"task":"Common Sense Reasoning","parent_task":null,"dataset":"PARus","model_name":"YaLM 1.0B few-shot","metrics":{"Accuracy":"0.766"},"paper_url":null,"paper_title":null,"paper_date":null,"code_links":[],"metrics_order":"[\"Accuracy\"]","area":"Natural Language Processing","uses_additional_data":0,"source":"archive","tags":[]},{"id":32199,"task":"Common Sense Reasoning","parent_task":null,"dataset":"PARus","model_name":"RuGPT3XL few-shot","metrics":{"Accuracy":"0.676"},"paper_url":null,"paper_title":null,"paper_date":null,"code_links":[],"metrics_order":"[\"Accuracy\"]","area":"Natural Language Processing","uses_additional_data":0,"source":"archive","tags":[]},{"id":32200,"task":"Common Sense Reasoning","parent_task":null,"dataset":"PARus","model_name":"ruT5-large-finetune","metrics":{"Accuracy":"0.66"},"paper_url":null,"paper_title":null,"paper_date":null,"code_links":[],"metrics_order":"[\"Accuracy\"]","area":"Natural Language Processing","uses_additional_data":0,"source":"archive","tags":[]},{"id":32201,"task":"Common Sense Reasoning","parent_task":null,"dataset":"PARus","model_name":"RuGPT3Medium","metrics":{"Accuracy":"0.598"},"paper_url":null,"paper_title":null,"paper_date":null,"code_links":[],"metrics_order":"[\"Accuracy\"]","area":"Natural Language Processing","uses_additional_data":0,"source":"archive","tags":[]},{"id":32202,"task":"Common Sense Reasoning","parent_task":null,"dataset":"PARus","model_name":"RuGPT3Large","metrics":{"Accuracy":"0.584"},"paper_url":null,"paper_title":null,"paper_date":null,"code_links":[],"metrics_order":"[\"Accuracy\"]","area":"Natural Language Processing","uses_additional_data":0,"source":"archive","tags":[]},{"id":32203,"task":"Common Sense Reasoning","parent_task":null,"dataset":"PARus","model_name":"RuBERT plain","metrics":{"Accuracy":"0.574"},"paper_url":null,"paper_title":null,"paper_date":null,"code_links":[],"metrics_order":"[\"Accuracy\"]","area":"Natural Language Processing","uses_additional_data":0,"source":"archive","tags":[]},{"id":32204,"task":"Common Sense Reasoning","parent_task":null,"dataset":"PARus","model_name":"RuGPT3Small","metrics":{"Accuracy":"0.562"},"paper_url":null,"paper_title":null,"paper_date":null,"code_links":[],"metrics_order":"[\"Accuracy\"]","area":"Natural Language Processing","uses_additional_data":0,"source":"archive","tags":[]},{"id":32205,"task":"Common Sense Reasoning","parent_task":null,"dataset":"PARus","model_name":"ruT5-base-finetune","metrics":{"Accuracy":"0.554"},"paper_url":null,"paper_title":null,"paper_date":null,"code_links":[],"metrics_order":"[\"Accuracy\"]","area":"Natural Language Processing","uses_additional_data":0,"source":"archive","tags":[]},{"id":32206,"task":"Common Sense Reasoning","parent_task":null,"dataset":"PARus","model_name":"Multilingual Bert","metrics":{"Accuracy":"0.528"},"paper_url":null,"paper_title":null,"paper_date":null,"code_links":[],"metrics_order":"[\"Accuracy\"]","area":"Natural Language Processing","uses_additional_data":0,"source":"archive","tags":[]},{"id":32207,"task":"Common Sense Reasoning","parent_task":null,"dataset":"PARus","model_name":"ruRoberta-large finetune","metrics":{"Accuracy":"0.508"},"paper_url":null,"paper_title":null,"paper_date":null,"code_links":[],"metrics_order":"[\"Accuracy\"]","area":"Natural Language Processing","uses_additional_data":0,"source":"archive","tags":[]},{"id":32208,"task":"Common Sense Reasoning","parent_task":null,"dataset":"PARus","model_name":"RuBERT conversational","metrics":{"Accuracy":"0.508"},"paper_url":null,"paper_title":null,"paper_date":null,"code_links":[],"metrics_order":"[\"Accuracy\"]","area":"Natural Language Processing","uses_additional_data":0,"source":"archive","tags":[]},{"id":32209,"task":"Common Sense Reasoning","parent_task":null,"dataset":"PARus","model_name":"MT5 Large","metrics":{"Accuracy":"0.504"},"paper_url":"https://arxiv.org/abs/2010.11934v3","paper_title":"mT5: A massively multilingual pre-trained text-to-text transformer","paper_date":"2020-10-22","code_links":[{"title":"huggingface/transformers","url":"https://github.com/huggingface/transformers"},{"title":"google-research/multilingual-t5","url":"https://github.com/google-research/multilingual-t5"},{"title":"google-research/byt5","url":"https://github.com/google-research/byt5"},{"title":"MorenoLaQuatra/bart-it","url":"https://github.com/MorenoLaQuatra/bart-it"},{"title":"manshri/tesum","url":"https://github.com/manshri/tesum"},{"title":"KoshiroSato/Flask_NLP_App","url":"https://github.com/KoshiroSato/Flask_NLP_App"},{"title":"pwc-1/Paper-5","url":"https://github.com/pwc-1/Paper-5/tree/main/mt5"},{"title":"KoshiroSato/Simple_Transformers_mT5_finetuning","url":"https://github.com/KoshiroSato/Simple_Transformers_mT5_finetuning"}],"metrics_order":"[\"Accuracy\"]","area":"Natural Language Processing","uses_additional_data":0,"source":"archive","tags":[]},{"id":32210,"task":"Common Sense Reasoning","parent_task":null,"dataset":"PARus","model_name":"SBERT_Large_mt_ru_finetuning","metrics":{"Accuracy":"0.498"},"paper_url":null,"paper_title":null,"paper_date":null,"code_links":[],"metrics_order":"[\"Accuracy\"]","area":"Natural Language Processing","uses_additional_data":0,"source":"archive","tags":[]},{"id":32211,"task":"Common Sense Reasoning","parent_task":null,"dataset":"PARus","model_name":"SBERT_Large","metrics":{"Accuracy":"0.498"},"paper_url":null,"paper_title":null,"paper_date":null,"code_links":[],"metrics_order":"[\"Accuracy\"]","area":"Natural Language Processing","uses_additional_data":0,"source":"archive","tags":[]},{"id":32212,"task":"Common Sense Reasoning","parent_task":null,"dataset":"PARus","model_name":"majority_class","metrics":{"Accuracy":"0.498"},"paper_url":"https://arxiv.org/abs/2105.01192v1","paper_title":"Unreasonable Effectiveness of Rule-Based Heuristics in Solving Russian SuperGLUE Tasks","paper_date":"2021-05-03","code_links":[],"metrics_order":"[\"Accuracy\"]","area":"Natural Language Processing","uses_additional_data":0,"source":"archive","tags":[]},{"id":32213,"task":"Common Sense Reasoning","parent_task":null,"dataset":"PARus","model_name":"ruBert-large finetune","metrics":{"Accuracy":"0.492"},"paper_url":null,"paper_title":null,"paper_date":null,"code_links":[],"metrics_order":"[\"Accuracy\"]","area":"Natural Language Processing","uses_additional_data":0,"source":"archive","tags":[]},{"id":32214,"task":"Common Sense Reasoning","parent_task":null,"dataset":"PARus","model_name":"Baseline TF-IDF1.1","metrics":{"Accuracy":"0.486"},"paper_url":"https://arxiv.org/abs/2010.15925v2","paper_title":"RussianSuperGLUE: A Russian Language Understanding Evaluation Benchmark","paper_date":"2020-10-29","code_links":[{"title":"RussianNLP/RussianSuperGLUE","url":"https://github.com/RussianNLP/RussianSuperGLUE"},{"title":"RussianNLP/MOROCCO","url":"https://github.com/RussianNLP/MOROCCO"}],"metrics_order":"[\"Accuracy\"]","area":"Natural Language Processing","uses_additional_data":0,"source":"archive","tags":[]},{"id":32215,"task":"Common Sense Reasoning","parent_task":null,"dataset":"PARus","model_name":"Random weighted","metrics":{"Accuracy":"0.48"},"paper_url":"https://arxiv.org/abs/2105.01192v1","paper_title":"Unreasonable Effectiveness of Rule-Based Heuristics in Solving Russian SuperGLUE Tasks","paper_date":"2021-05-03","code_links":[],"metrics_order":"[\"Accuracy\"]","area":"Natural Language Processing","uses_additional_data":0,"source":"archive","tags":[]},{"id":32216,"task":"Common Sense Reasoning","parent_task":null,"dataset":"PARus","model_name":"heuristic majority","metrics":{"Accuracy":"0.478"},"paper_url":"https://arxiv.org/abs/2105.01192v1","paper_title":"Unreasonable Effectiveness of Rule-Based Heuristics in Solving Russian SuperGLUE Tasks","paper_date":"2021-05-03","code_links":[],"metrics_order":"[\"Accuracy\"]","area":"Natural Language Processing","uses_additional_data":0,"source":"archive","tags":[]},{"id":32217,"task":"Common Sense Reasoning","parent_task":null,"dataset":"PARus","model_name":"ruBert-base finetune","metrics":{"Accuracy":"0.476"},"paper_url":null,"paper_title":null,"paper_date":null,"code_links":[],"metrics_order":"[\"Accuracy\"]","area":"Natural Language Processing","uses_additional_data":0,"source":"archive","tags":[]}]}