{"paper":{"title":"Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"Sentence-BERT uses siamese and triplet training on BERT to create fixed sentence embeddings that support fast cosine-similarity comparisons while matching original accuracy.","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Iryna Gurevych, Nils Reimers","submitted_at":"2019-08-27T08:50:17Z","abstract_excerpt":"BERT (Devlin et al., 2018) and RoBERTa (Liu et al., 2019) has set a new state-of-the-art performance on sentence-pair regression tasks like semantic textual similarity (STS). However, it requires that both sentences are fed into the network, which causes a massive computational overhead: Finding the most similar pair in a collection of 10,000 sentences requires about 50 million inference computations (~65 hours) with BERT. The construction of BERT makes it unsuitable for semantic similarity search as well as for unsupervised tasks like clustering.\n  In this publication, we present Sentence-BER"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"we present Sentence-BERT (SBERT), a modification of the pretrained BERT network that use siamese and triplet network structures to derive semantically meaningful sentence embeddings that can be compared using cosine-similarity. This reduces the effort for finding the most similar pair from 65 hours with BERT / RoBERTa to about 5 seconds with SBERT, while maintaining the accuracy from BERT.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That fine-tuning BERT with siamese and triplet networks produces standalone sentence embeddings whose cosine similarities accurately reflect semantic similarity at the level of the original pairwise BERT inference.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"Sentence-BERT adapts BERT with siamese and triplet networks to produce sentence embeddings for efficient cosine-similarity comparisons, cutting computation time from hours to seconds on similarity search while matching BERT accuracy.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"Sentence-BERT uses siamese and triplet training on BERT to create fixed sentence embeddings that support fast cosine-similarity comparisons while matching original accuracy.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"eab657b24b93e83c3e3307fb58ddc9c0febd889486866a1e63f45cb764981d20"},"source":{"id":"1908.10084","kind":"arxiv","version":1},"verdict":{"id":"282d487d-d305-4302-9665-5b865e4d22e6","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-10T14:47:49.156617Z","strongest_claim":"we present Sentence-BERT (SBERT), a modification of the pretrained BERT network that use siamese and triplet network structures to derive semantically meaningful sentence embeddings that can be compared using cosine-similarity. This reduces the effort for finding the most similar pair from 65 hours with BERT / RoBERTa to about 5 seconds with SBERT, while maintaining the accuracy from BERT.","one_line_summary":"Sentence-BERT adapts BERT with siamese and triplet networks to produce sentence embeddings for efficient cosine-similarity comparisons, cutting computation time from hours to seconds on similarity search while matching BERT accuracy.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That fine-tuning BERT with siamese and triplet networks produces standalone sentence embeddings whose cosine similarities accurately reflect semantic similarity at the level of the original pairwise BERT inference.","pith_extraction_headline":"Sentence-BERT uses siamese and triplet training on BERT to create fixed sentence embeddings that support fast cosine-similarity comparisons while matching original accuracy."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/1908.10084/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":38,"sample":[{"doi":"","year":2015,"title":"Eneko Agirre, Carmen Banea, Claire Cardie, Daniel Cer, Mona Diab, Aitor Gonzalez-Agirre, Weiwei Guo, Inigo Lopez-Gazpio, Montse Maritxalar, Rada Mihalcea, German Rigau, Larraitz Uria, and Janyce Wiebe","work_id":"045e6092-8c74-4f10-a95a-5f99899e9866","ref_index":1,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"10.3115/v1/s14-2010","year":2014,"title":"Eneko Agirre, Carmen Banea, Claire Cardie, Daniel Cer, Mona Diab, Aitor Gonzalez-Agirre, Weiwei Guo, Rada Mihalcea, German Rigau, and Janyce Wiebe. 2014. https://doi.org/10.3115/v1/S14-2010 S em E val","work_id":"9644ab35-855a-4f87-97a6-b7db05cb97f9","ref_index":2,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2016,"title":"Eneko Agirre, Carmen Banea, Daniel M. Cer, Mona T. Diab, Aitor Gonzalez - Agirre, Rada Mihalcea, German Rigau, and Janyce Wiebe. 2016. http://aclweb.org/anthology/S/S16/S16-1081.pdf SemEval-2016 Task ","work_id":"b53a638c-aa98-4acd-b768-2150a7c9b7ed","ref_index":3,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2013,"title":"Eneko Agirre, Daniel Cer, Mona Diab, Aitor Gonzalez-Agirre, and Weiwei Guo. 2013. https://www.aclweb.org/anthology/S13-1004 * SEM 2013 shared task: Semantic Textual Similarity . In Second Joint Confer","work_id":"73f9adc0-1e66-417a-94b8-b2554aa4e341","ref_index":4,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2012,"title":"Eneko Agirre, Mona Diab, Daniel Cer, and Aitor Gonzalez-Agirre. 2012. http://dl.acm.org/citation.cfm?id=2387636.2387697 SemEval-2012 Task 6: A Pilot on Semantic Textual Similarity . In Proceedings of ","work_id":"b69745b4-4076-4c10-b7a6-fce20eb44197","ref_index":5,"cited_arxiv_id":"","is_internal_anchor":false}],"resolved_work":38,"snapshot_sha256":"04424c6fbd6ae4d1d05787feaf8b6b0e5267a9420979d64eb26d1177dde4a201","internal_anchors":5},"formal_canon":{"evidence_count":2,"snapshot_sha256":"b57db49ec398ecfe15cc90a5ac76220a8c8f98935943645c847ddd6a1e4581db"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"}