{"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/flaubert-unsupervised-language-model-pre","title":"FlauBERT: Unsupervised Language Model Pre-training for French","arxiv_id":"1912.05372","date":"2019-12-11","proceeding":"LREC 2020 5","authors":["Hang Le","Loïc Vial","Jibril Frej","Vincent Segonne","Maximin Coavoux","Benjamin Lecouteux","Alexandre Allauzen","Benoît Crabbé","Laurent Besacier","Didier Schwab"],"abstract":"Language models have become a key step to achieve state-of-the art results in many different Natural Language Processing (NLP) tasks. Leveraging the huge amount of unlabeled texts nowadays available, they provide an efficient way to pre-train continuous word representations that can be fine-tuned for a downstream task, along with their contextualization at the sentence level. This has been widely demonstrated for English using contextualized representations (Dai and Le, 2015; Peters et al., 2018; Howard and Ruder, 2018; Radford et al., 2018; Devlin et al., 2019; Yang et al., 2019b). In this paper, we introduce and share FlauBERT, a model learned on a very large and heterogeneous French corpus. Models of different sizes are trained using the new CNRS (French National Centre for Scientific Research) Jean Zay supercomputer. We apply our French language models to diverse NLP tasks (text classification, paraphrasing, natural language inference, parsing, word sense disambiguation) and show that most of the time they outperform other pre-training approaches. Different versions of FlauBERT as well as a unified evaluation protocol for the downstream tasks, called FLUE (French Language Understanding Evaluation), are shared to the research community for further reproducible experiments in French NLP.","url_abs":"https://arxiv.org/abs/1912.05372v4","url_pdf":"https://arxiv.org/pdf/1912.05372v4.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":"flaubert-unsupervised-language-model-pre","repo_url":"https://github.com/getalp/Flaubert","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"NOASSERTION"}},{"paper_slug":"flaubert-unsupervised-language-model-pre","repo_url":"https://github.com/TheophileBlard/french-sentiment-analysis-with-bert","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":{"status":"ok","spdx":"MIT"}},{"paper_slug":"flaubert-unsupervised-language-model-pre","repo_url":"https://github.com/bencrabbe/npdependency","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"NOASSERTION"}},{"paper_slug":"flaubert-unsupervised-language-model-pre","repo_url":"https://github.com/bourrel/French-News-Clustering","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":{"status":"ok"}},{"paper_slug":"flaubert-unsupervised-language-model-pre","repo_url":"https://github.com/getalp/disambiguate","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}},{"paper_slug":"flaubert-unsupervised-language-model-pre","repo_url":"https://github.com/huggingface/transformers","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"flaubert-unsupervised-language-model-pre","repo_url":"https://github.com/ialifinaritra/text_summarization","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}}],"tasks":[{"task_slug":"flue","task_name":"FLUE"},{"task_slug":"language-modeling","task_name":"Language Modeling"},{"task_slug":"language-modelling","task_name":"Language Modelling"},{"task_slug":"natural-language-inference","task_name":"Natural Language Inference"},{"task_slug":"sentence","task_name":"Sentence"},{"task_slug":"text-classification","task_name":"Text Classification"},{"task_slug":"word-sense-disambiguation","task_name":"Word Sense Disambiguation"},{"task_slug":"model","task_name":"model"}],"methods":[],"datasets_introduced":[{"slug":"flue-french-language-understanding-evaluation","name":"FLUE","full_name":"French Language Understanding Evaluation"}],"methods_introduced":[],"results":[{"leaderboard":"/sota/natural-language-inference-on-xnli-french","task":"Natural Language Inference","dataset":"XNLI French","model":"FlauBERT (large)","rank_in_archive_order":2,"of":6,"metrics":{"Accuracy":"83.4"},"uses_additional_data":false},{"leaderboard":"/sota/natural-language-inference-on-xnli-french","task":"Natural Language Inference","dataset":"XNLI French","model":"FlauBERT (base)","rank_in_archive_order":4,"of":6,"metrics":{"Accuracy":"80.6"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1912.05372","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"1912.05372"}},"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/huggingface/transformers","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/bencrabbe/npdependency","reach":{"status":"ok","spdx":"NOASSERTION"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/getalp/disambiguate","reach":{"status":"ok","spdx":"MIT"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/ialifinaritra/text_summarization","reach":{"status":"ok","spdx":"MIT"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/TheophileBlard/french-sentiment-analysis-with-bert","reach":{"status":"ok","spdx":"MIT"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/getalp/Flaubert","reach":{"status":"ok","spdx":"NOASSERTION"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/bourrel/French-News-Clustering","reach":{"status":"ok"}}],"summary":{"ran_violates":1,"unverified":8},"by_repo_kind":{"listed":{"samples":9,"ran":1,"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":0,"samples":[{"code_sha256_prefix":"0c93830510f645a3","entry":"check_sru_requirement","repo":"getalp/disambiguate","repo_kind":"listed","path":"python/onmt/models/sru.py","file_url":"https://github.com/getalp/disambiguate/blob/HEAD/python/onmt/models/sru.py","link_basis":"harvester_set","language":"python","status":"ran_violates","verification_level":1,"contract_check":"VIOLATES","metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"0c93830510f645a3"}},{"code_sha256_prefix":"b60aff23d14fc8ae","entry":"collapse_copy_scores","repo":"getalp/disambiguate","repo_kind":"listed","path":"python/onmt/modules/copy_generator.py","file_url":"https://github.com/getalp/disambiguate/blob/HEAD/python/onmt/modules/copy_generator.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":"b60aff23d14fc8ae"}},{"code_sha256_prefix":"655c4a8978903a61","entry":"context_gate_factory","repo":"getalp/disambiguate","repo_kind":"listed","path":"python/onmt/modules/gate.py","file_url":"https://github.com/getalp/disambiguate/blob/HEAD/python/onmt/modules/gate.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":"655c4a8978903a61"}},{"code_sha256_prefix":"1428b5516b0a164b","entry":"get_data","repo":"TheophileBlard/french-sentiment-analysis-with-bert","repo_kind":"listed","path":"utils_acl.py","file_url":"https://github.com/TheophileBlard/french-sentiment-analysis-with-bert/blob/HEAD/utils_acl.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":"1428b5516b0a164b"}},{"code_sha256_prefix":"7843e5379321e72d","entry":"parse_file","repo":"TheophileBlard/french-sentiment-analysis-with-bert","repo_kind":"listed","path":"utils_acl.py","file_url":"https://github.com/TheophileBlard/french-sentiment-analysis-with-bert/blob/HEAD/utils_acl.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":"7843e5379321e72d"}},{"code_sha256_prefix":"3ba692edead757df","entry":"plot_training_curves","repo":"TheophileBlard/french-sentiment-analysis-with-bert","repo_kind":"listed","path":"utils.py","file_url":"https://github.com/TheophileBlard/french-sentiment-analysis-with-bert/blob/HEAD/utils.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":"3ba692edead757df"}},{"code_sha256_prefix":"3766be0864eb80f0","entry":"print_confusion_matrix","repo":"TheophileBlard/french-sentiment-analysis-with-bert","repo_kind":"listed","path":"utils.py","file_url":"https://github.com/TheophileBlard/french-sentiment-analysis-with-bert/blob/HEAD/utils.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":"3766be0864eb80f0"}},{"code_sha256_prefix":"fbdb4336ee093312","entry":"process_item","repo":"TheophileBlard/french-sentiment-analysis-with-bert","repo_kind":"listed","path":"utils_acl.py","file_url":"https://github.com/TheophileBlard/french-sentiment-analysis-with-bert/blob/HEAD/utils_acl.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":"fbdb4336ee093312"}},{"code_sha256_prefix":"0c3bddd7d3eaba02","entry":"seq_linear","repo":"getalp/disambiguate","repo_kind":"listed","path":"python/onmt/modules/conv_multi_step_attention.py","file_url":"https://github.com/getalp/disambiguate/blob/HEAD/python/onmt/modules/conv_multi_step_attention.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":"0c3bddd7d3eaba02"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}