{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/glue-a-multi-task-benchmark-and-analysis","title":"GLUE: A Multi-Task Benchmark and Analysis Platform for Natural Language Understanding","arxiv_id":"1804.07461","date":"2018-04-20","proceeding":"WS 2018 11","authors":["Alex Wang","Amanpreet Singh","Julian Michael","Felix Hill","Omer Levy","Samuel R. Bowman"],"abstract":"For natural language understanding (NLU) technology to be maximally useful,\nboth practically and as a scientific object of study, it must be general: it\nmust be able to process language in a way that is not exclusively tailored to\nany one specific task or dataset. In pursuit of this objective, we introduce\nthe General Language Understanding Evaluation benchmark (GLUE), a tool for\nevaluating and analyzing the performance of models across a diverse range of\nexisting NLU tasks. GLUE is model-agnostic, but it incentivizes sharing\nknowledge across tasks because certain tasks have very limited training data.\nWe further provide a hand-crafted diagnostic test suite that enables detailed\nlinguistic analysis of NLU models. We evaluate baselines based on current\nmethods for multi-task and transfer learning and find that they do not\nimmediately give substantial improvements over the aggregate performance of\ntraining a separate model per task, indicating room for improvement in\ndeveloping general and robust NLU systems.","url_abs":"http://arxiv.org/abs/1804.07461v3","url_pdf":"http://arxiv.org/pdf/1804.07461v3.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"glue-a-multi-task-benchmark-and-analysis","repo_url":"https://github.com/jsalt18-sentence-repl/jiant","is_official":0,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}},{"paper_slug":"glue-a-multi-task-benchmark-and-analysis","repo_url":"https://github.com/nyu-mll/GLUE-baselines","is_official":0,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"pytorch","reach":null},{"paper_slug":"glue-a-multi-task-benchmark-and-analysis","repo_url":"https://github.com/ashi-ta/speechglue","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok","spdx":"NOASSERTION"}},{"paper_slug":"glue-a-multi-task-benchmark-and-analysis","repo_url":"https://github.com/benzakenelad/BitFit","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}},{"paper_slug":"glue-a-multi-task-benchmark-and-analysis","repo_url":"https://github.com/colinzhaoust/intrinsic_fewshot_hardness","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok","spdx":"MIT"}},{"paper_slug":"glue-a-multi-task-benchmark-and-analysis","repo_url":"https://github.com/nvshrao/Pytorch-GLUE","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok"}},{"paper_slug":"glue-a-multi-task-benchmark-and-analysis","repo_url":"https://github.com/smallbenchnlp/benchmark","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok"}},{"paper_slug":"glue-a-multi-task-benchmark-and-analysis","repo_url":"https://github.com/alibaba/EasyNLP","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"jax","reach":{"status":"ok","spdx":"Apache-2.0"}},{"paper_slug":"glue-a-multi-task-benchmark-and-analysis","repo_url":"https://github.com/imran3180/pytorch-nli","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"pytorch","reach":{"status":"ok"}},{"paper_slug":"glue-a-multi-task-benchmark-and-analysis","repo_url":"https://github.com/kainoj/run_glue","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"pytorch","reach":{"status":"ok"}},{"paper_slug":"glue-a-multi-task-benchmark-and-analysis","repo_url":"https://github.com/ofa-sys/ofa","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"pytorch","reach":{"status":"ok","spdx":"Apache-2.0"}}],"tasks":[{"task_slug":"diagnostic","task_name":"Diagnostic"},{"task_slug":"natural-language-inference","task_name":"Natural Language Inference"},{"task_slug":"natural-language-understanding","task_name":"Natural Language Understanding"},{"task_slug":"qqp","task_name":"QQP"},{"task_slug":"transfer-learning","task_name":"Transfer Learning"}],"methods":[],"datasets_introduced":[{"slug":"glue","name":"GLUE","full_name":"General Language Understanding Evaluation benchmark"},{"slug":"qnli","name":"QNLI","full_name":"Question-answering NLI"}],"methods_introduced":[],"results":[{"leaderboard":"/sota/natural-language-inference-on-multinli","task":"Natural Language Inference","dataset":"MultiNLI","model":"Multi-task BiLSTM + Attn","rank_in_archive_order":49,"of":67,"metrics":{"Matched":"72.2","Mismatched":"72.1"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1804.07461","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"1804.07461"}},"developers":"https://syntology.ai/developers","read_at":"2026-09-24T18:15:14+00:00","read_at_is":"when the build read Syntology's graph, not when any sample ran","claim":"Per-sample execution status on synthesized fixtures; not a correctness claim about the paper. Samples come from repositories linked to the paper, official or community; repo_kind says which.","repos":[{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/ashi-ta/speechglue","reach":{"status":"ok","spdx":"NOASSERTION"}},{"provenance":"deterministic:regex_extraction","url":"https://github.com/jsalt18-sentence-repl/jiant","reach":{"status":"ok","spdx":"MIT"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/benzakenelad/BitFit","reach":{"status":"ok","spdx":"MIT"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/imran3180/pytorch-nli","reach":{"status":"ok"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/nvshrao/Pytorch-GLUE","reach":{"status":"ok"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/ofa-sys/ofa","reach":{"status":"ok","spdx":"Apache-2.0"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/kainoj/run_glue","reach":{"status":"ok"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/smallbenchnlp/benchmark","reach":{"status":"ok"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/alibaba/EasyNLP","reach":{"status":"ok","spdx":"Apache-2.0"}},{"provenance":"deterministic:regex_extraction","url":"https://github.com/nyu-mll/GLUE-baselines","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/colinzhaoust/intrinsic_fewshot_hardness","reach":{"status":"ok","spdx":"MIT"}}],"summary":{"ran_honours":1,"ran_draft_wrong":2,"ran":2,"unverified":9},"by_repo_kind":{"named_in_paper":{"samples":14,"ran":5,"repositories":2}},"repo_kind_vocabulary":{"official":"The archive marks this repository official for the paper","named_in_paper":"The archive records that the paper mentions this repository; it is not marked official","listed":"In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper","found_in_text":"Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted","community":"Not in the archive's code links for this paper; a community repository Syntology harvested"},"n_pointer_only_for_licence":1,"samples":[{"code_sha256_prefix":"8f7619b2828926ea","entry":"flat_collate_fn","repo":"jsalt18-sentence-repl/jiant","repo_kind":"named_in_paper","path":"jiant/tasks/core.py","file_url":"https://github.com/jsalt18-sentence-repl/jiant/blob/HEAD/jiant/tasks/core.py","link_basis":"plan_row","language":"python","status":"ran_honours","verification_level":1,"contract_check":"HONOURS","metamorphic_tier":"well_formed","behaviour_fingerprint":true,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"8f7619b2828926ea"}},{"code_sha256_prefix":"3caf26b311c98658","entry":"get_tasks","repo":"nyu-mll/GLUE-baselines","repo_kind":"named_in_paper","path":"download_glue_data.py","file_url":"https://github.com/nyu-mll/GLUE-baselines/blob/HEAD/download_glue_data.py","link_basis":"first_harvest_node","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"OUTPUT_MISDECLARED","metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"3caf26b311c98658"}},{"code_sha256_prefix":"61ffc5549a75fe3c","entry":"metadata_collate_fn","repo":"jsalt18-sentence-repl/jiant","repo_kind":"named_in_paper","path":"jiant/tasks/core.py","file_url":"https://github.com/jsalt18-sentence-repl/jiant/blob/HEAD/jiant/tasks/core.py","link_basis":"harvester_set","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"OUTPUT_MISDECLARED","metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"61ffc5549a75fe3c"}},{"code_sha256_prefix":"e17e83a5650fd73c","entry":"resolve_is_lower_case","repo":"jsalt18-sentence-repl/jiant","repo_kind":"named_in_paper","path":"jiant/shared/model_resolution.py","file_url":"https://github.com/jsalt18-sentence-repl/jiant/blob/HEAD/jiant/shared/model_resolution.py","link_basis":"plan_row","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"e17e83a5650fd73c"}},{"code_sha256_prefix":"36c8a80131f2e20a","entry":"resolve_warmup_steps","repo":"jsalt18-sentence-repl/jiant","repo_kind":"named_in_paper","path":"jiant/shared/model_setup.py","file_url":"https://github.com/jsalt18-sentence-repl/jiant/blob/HEAD/jiant/shared/model_setup.py","link_basis":"plan_row","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"36c8a80131f2e20a"}},{"code_sha256_prefix":"e445a3b7d6a58154","entry":"batched_index_select","repo":"jsalt18-sentence-repl/jiant","repo_kind":"named_in_paper","path":"jiant/ext/allennlp.py","file_url":"https://github.com/jsalt18-sentence-repl/jiant/blob/HEAD/jiant/ext/allennlp.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"e445a3b7d6a58154"}},{"code_sha256_prefix":"eaa6b3cef2ce9444","entry":"compare_dataset_with_metadata","repo":"jsalt18-sentence-repl/jiant","repo_kind":"named_in_paper","path":"jiant/shared/caching.py","file_url":"https://github.com/jsalt18-sentence-repl/jiant/blob/HEAD/jiant/shared/caching.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"eaa6b3cef2ce9444"}},{"code_sha256_prefix":"0f5c37217af75b26","entry":"compare_tensor_tuples","repo":"jsalt18-sentence-repl/jiant","repo_kind":"named_in_paper","path":"jiant/shared/caching.py","file_url":"https://github.com/jsalt18-sentence-repl/jiant/blob/HEAD/jiant/shared/caching.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"0f5c37217af75b26"}},{"code_sha256_prefix":"9b5f1ac1298e3440","entry":"complex_backpropagate","repo":"jsalt18-sentence-repl/jiant","repo_kind":"named_in_paper","path":"jiant/shared/runner.py","file_url":"https://github.com/jsalt18-sentence-repl/jiant/blob/HEAD/jiant/shared/runner.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"9b5f1ac1298e3440"}},{"code_sha256_prefix":"b91635f00cdc7b71","entry":"convert_to_chunks","repo":"jsalt18-sentence-repl/jiant","repo_kind":"named_in_paper","path":"jiant/shared/caching.py","file_url":"https://github.com/jsalt18-sentence-repl/jiant/blob/HEAD/jiant/shared/caching.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"b91635f00cdc7b71"}},{"code_sha256_prefix":"fa0fcf05318dae3d","entry":"init_cuda_from_args","repo":"jsalt18-sentence-repl/jiant","repo_kind":"named_in_paper","path":"jiant/shared/initialization.py","file_url":"https://github.com/jsalt18-sentence-repl/jiant/blob/HEAD/jiant/shared/initialization.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"fa0fcf05318dae3d"}},{"code_sha256_prefix":"540028e9d806d8a8","entry":"init_seed","repo":"jsalt18-sentence-repl/jiant","repo_kind":"named_in_paper","path":"jiant/shared/initialization.py","file_url":"https://github.com/jsalt18-sentence-repl/jiant/blob/HEAD/jiant/shared/initialization.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"540028e9d806d8a8"}},{"code_sha256_prefix":"16cde5ff8ae0a586","entry":"masked_softmax","repo":"jsalt18-sentence-repl/jiant","repo_kind":"named_in_paper","path":"jiant/ext/allennlp.py","file_url":"https://github.com/jsalt18-sentence-repl/jiant/blob/HEAD/jiant/ext/allennlp.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"16cde5ff8ae0a586"}},{"code_sha256_prefix":"2f417d486bcad523","entry":"weighted_sum","repo":"jsalt18-sentence-repl/jiant","repo_kind":"named_in_paper","path":"jiant/ext/allennlp.py","file_url":"https://github.com/jsalt18-sentence-repl/jiant/blob/HEAD/jiant/ext/allennlp.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"2f417d486bcad523"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}