{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:GHP3J7IEEKG3CJSUG2H7D2RWKB","short_pith_number":"pith:GHP3J7IE","schema_version":"1.0","canonical_sha256":"31dfb4fd04228db12654368ff1ea36506f644e36f676c6f5d3fa40464510e075","source":{"kind":"arxiv","id":"2103.15316","version":1},"attestation_state":"computed","paper":{"title":"Whitening Sentence Representations for Better Semantics and Faster Retrieval","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CL","authors_text":"Jianlin Su, Jiarun Cao, Weijie Liu, Yangyiwen Ou","submitted_at":"2021-03-29T03:51:53Z","abstract_excerpt":"Pre-training models such as BERT have achieved great success in many natural language processing tasks. However, how to obtain better sentence representation through these pre-training models is still worthy to exploit. Previous work has shown that the anisotropy problem is an critical bottleneck for BERT-based sentence representation which hinders the model to fully utilize the underlying semantic features. Therefore, some attempts of boosting the isotropy of sentence distribution, such as flow-based model, have been applied to sentence representations and achieved some improvement. In this p"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2103.15316","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.CL","submitted_at":"2021-03-29T03:51:53Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"12ce417c2a71d0880a2ca40b020cbebe5febcaff60dceed30cc72b5bde8f1254","abstract_canon_sha256":"6af92348a039d44b2e0ca57ea8435e72f3b236bed0b1999eac4d43c50999af94"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T02:27:04.724662Z","signature_b64":"dc124s8UUCdiSBde94Yysjz0Qfu739G8NErbaCcNGBDSBI8zqPDeY9DdraSlc4YEpt6qnJnE45hJE8LE8/PSAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"31dfb4fd04228db12654368ff1ea36506f644e36f676c6f5d3fa40464510e075","last_reissued_at":"2026-07-05T02:27:04.724108Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T02:27:04.724108Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Whitening Sentence Representations for Better Semantics and Faster Retrieval","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CL","authors_text":"Jianlin Su, Jiarun Cao, Weijie Liu, Yangyiwen Ou","submitted_at":"2021-03-29T03:51:53Z","abstract_excerpt":"Pre-training models such as BERT have achieved great success in many natural language processing tasks. However, how to obtain better sentence representation through these pre-training models is still worthy to exploit. Previous work has shown that the anisotropy problem is an critical bottleneck for BERT-based sentence representation which hinders the model to fully utilize the underlying semantic features. Therefore, some attempts of boosting the isotropy of sentence distribution, such as flow-based model, have been applied to sentence representations and achieved some improvement. In this p"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2103.15316","kind":"arxiv","version":1},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2103.15316/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"aliases":[{"alias_kind":"arxiv","alias_value":"2103.15316","created_at":"2026-07-05T02:27:04.724181+00:00"},{"alias_kind":"arxiv_version","alias_value":"2103.15316v1","created_at":"2026-07-05T02:27:04.724181+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2103.15316","created_at":"2026-07-05T02:27:04.724181+00:00"},{"alias_kind":"pith_short_12","alias_value":"GHP3J7IEEKG3","created_at":"2026-07-05T02:27:04.724181+00:00"},{"alias_kind":"pith_short_16","alias_value":"GHP3J7IEEKG3CJSU","created_at":"2026-07-05T02:27:04.724181+00:00"},{"alias_kind":"pith_short_8","alias_value":"GHP3J7IE","created_at":"2026-07-05T02:27:04.724181+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":18,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2607.07159","citing_title":"Recovering Latent Structures after Variational Bayesian Variable Selection: Fit Assessment and Factor-Number Selection in Partially Exploratory Factor Analysis","ref_index":196,"is_internal_anchor":true},{"citing_arxiv_id":"2606.25272","citing_title":"Semantic Code Clone Detection: Are We There Yet?","ref_index":38,"is_internal_anchor":false},{"citing_arxiv_id":"2606.24937","citing_title":"The Hitchhiker's Guide to Agentic AI: From Foundations to Systems","ref_index":41,"is_internal_anchor":false},{"citing_arxiv_id":"2606.19692","citing_title":"When Global Gating Is Enough: Admission-Time Hubness Control in Anisotropic Vector Retrieval Systems","ref_index":27,"is_internal_anchor":false},{"citing_arxiv_id":"2606.17449","citing_title":"MODE-RAG: Manifold Outlier Diagnosis and Energy-based Retrieval-Augmented Generation Evaluation","ref_index":48,"is_internal_anchor":false},{"citing_arxiv_id":"2606.12793","citing_title":"Semantic Identification of IoT Devices from Behavioral Primitives","ref_index":27,"is_internal_anchor":false},{"citing_arxiv_id":"2606.09672","citing_title":"Correlation Is Not Enough: Embedding Human Metadata for Individual Causal Discovery","ref_index":4,"is_internal_anchor":false},{"citing_arxiv_id":"2606.07502","citing_title":"Your UnEmbedding Matrix is Secretly a Feature Lens for Text Embeddings","ref_index":29,"is_internal_anchor":false},{"citing_arxiv_id":"2606.29571","citing_title":"Anisotropy Decides Cosine vs. Rank Metrics for Text Embeddings","ref_index":20,"is_internal_anchor":false},{"citing_arxiv_id":"2605.29286","citing_title":"CrossAlpha: An Annual-Report Benchmark for Cross-Market Factor Researc (with LLM Agents)","ref_index":5,"is_internal_anchor":false},{"citing_arxiv_id":"2605.18226","citing_title":"Context Memorization for Efficient Long Context Generation","ref_index":38,"is_internal_anchor":false},{"citing_arxiv_id":"2201.10005","citing_title":"Text and Code Embeddings by Contrastive Pre-Training","ref_index":22,"is_internal_anchor":false},{"citing_arxiv_id":"2603.19339","citing_title":"Spectral Tempering for Embedding Compression in Dense Passage Retrieval","ref_index":37,"is_internal_anchor":false},{"citing_arxiv_id":"2104.08821","citing_title":"SimCSE: Simple Contrastive Learning of Sentence Embeddings","ref_index":64,"is_internal_anchor":false},{"citing_arxiv_id":"2308.03281","citing_title":"Towards General Text Embeddings with Multi-stage Contrastive Learning","ref_index":46,"is_internal_anchor":false},{"citing_arxiv_id":"2604.22549","citing_title":"ASPIRE: Make Spectral Graph Collaborative Filtering Great Again via Adaptive Filter Learning","ref_index":46,"is_internal_anchor":false},{"citing_arxiv_id":"2605.01609","citing_title":"Concepts Whisper While Syntax Shouts: Spectral Anti-Concentration and the Dual Geometry of Transformer Representations","ref_index":41,"is_internal_anchor":false},{"citing_arxiv_id":"2604.17257","citing_title":"REZE: Representation Regularization for Domain-adaptive Text Embedding Pre-finetuning","ref_index":62,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/GHP3J7IEEKG3CJSUG2H7D2RWKB","json":"https://pith.science/pith/GHP3J7IEEKG3CJSUG2H7D2RWKB.json","graph_json":"https://pith.science/api/pith-number/GHP3J7IEEKG3CJSUG2H7D2RWKB/graph.json","events_json":"https://pith.science/api/pith-number/GHP3J7IEEKG3CJSUG2H7D2RWKB/events.json","paper":"https://pith.science/paper/GHP3J7IE"},"agent_actions":{"view_html":"https://pith.science/pith/GHP3J7IEEKG3CJSUG2H7D2RWKB","download_json":"https://pith.science/pith/GHP3J7IEEKG3CJSUG2H7D2RWKB.json","view_paper":"https://pith.science/paper/GHP3J7IE","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2103.15316&json=true","fetch_graph":"https://pith.science/api/pith-number/GHP3J7IEEKG3CJSUG2H7D2RWKB/graph.json","fetch_events":"https://pith.science/api/pith-number/GHP3J7IEEKG3CJSUG2H7D2RWKB/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/GHP3J7IEEKG3CJSUG2H7D2RWKB/action/timestamp_anchor","attest_storage":"https://pith.science/pith/GHP3J7IEEKG3CJSUG2H7D2RWKB/action/storage_attestation","attest_author":"https://pith.science/pith/GHP3J7IEEKG3CJSUG2H7D2RWKB/action/author_attestation","sign_citation":"https://pith.science/pith/GHP3J7IEEKG3CJSUG2H7D2RWKB/action/citation_signature","submit_replication":"https://pith.science/pith/GHP3J7IEEKG3CJSUG2H7D2RWKB/action/replication_record"}},"created_at":"2026-07-05T02:27:04.724181+00:00","updated_at":"2026-07-05T02:27:04.724181+00:00"}