OpenCodePapers

natural-language-inference-on-lidirus

Natural Language Inference
Dataset Link
Results over time
Click legend items to toggle metrics. Hover points for model names.
Leaderboard
PaperCodeMCCModelNameReleaseDate
RussianSuperGLUE: A Russian Language Understanding Evaluation Benchmark✓ Link0.626Human Benchmark2020-10-29
0.339ruRoberta-large finetune
0.32ruT5-large-finetune
0.267ruT5-base-finetune
0.235ruBert-large finetune
0.231RuGPT3Large
0.224ruBert-base finetune
0.218SBERT_Large_mt_ru_finetuning
0.209SBERT_Large
0.191RuBERT plain
0.189Multilingual Bert
0.178RuBERT conversational
Unreasonable Effectiveness of Rule-Based Heuristics in Solving Russian SuperGLUE Tasks0.147heuristic majority2021-05-03
0.124YaLM 1.0B few-shot
0.096RuGPT3XL few-shot
mT5: A massively multilingual pre-trained text-to-text transformer✓ Link0.061MT5 Large2020-10-22
RussianSuperGLUE: A Russian Language Understanding Evaluation Benchmark✓ Link0.06Baseline TF-IDF1.12020-10-29
0.01RuGPT3Medium
0Golden Transformer
Unreasonable Effectiveness of Rule-Based Heuristics in Solving Russian SuperGLUE Tasks0Random weighted2021-05-03
Unreasonable Effectiveness of Rule-Based Heuristics in Solving Russian SuperGLUE Tasks0majority_class2021-05-03
-0.013RuGPT3Small