{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:ZGNGGPEP42CFM5MJF2DN3637OZ","short_pith_number":"pith:ZGNGGPEP","schema_version":"1.0","canonical_sha256":"c99a633c8fe6845675892e86ddfb7f764ed0c294c6d3642f8fd910c1e68fba08","source":{"kind":"arxiv","id":"2301.12005","version":2},"attestation_state":"computed","paper":{"title":"EmbedDistill: A Geometric Knowledge Distillation for Information Retrieval","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Aditya Krishna Menon, Ankit Singh Rawat, Manzil Zaheer, Rob Fergus, Sadeep Jayasumana, Sanjiv Kumar, Seungyeon Kim, Veeranjaneyulu Sadhanala, Wittawat Jitkrittum","submitted_at":"2023-01-27T22:04:37Z","abstract_excerpt":"Large neural models (such as Transformers) achieve state-of-the-art performance for information retrieval (IR). In this paper, we aim to improve distillation methods that pave the way for the resource-efficient deployment of such models in practice. Inspired by our theoretical analysis of the teacher-student generalization gap for IR models, we propose a novel distillation approach that leverages the relative geometry among queries and documents learned by the large teacher model. Unlike existing teacher score-based distillation methods, our proposed approach employs embedding matching tasks t"},"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":"2301.12005","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2023-01-27T22:04:37Z","cross_cats_sorted":[],"title_canon_sha256":"174b871a050b851fcbf7664e5e908ad5548e77833939361356541fd2484d8de9","abstract_canon_sha256":"1d6cffe627606d7a6aeda37375617b976126c6eec25b13b2e31c7f4713a576ef"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:27:45.375788Z","signature_b64":"+OBJAtXKa5MRi5hbsgfBuIoq1SWjRpl6lI3ufClcctinKGe2g3qmX9OMhyD6J+goCX0WCGDHefxvFOn+Z7EZCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c99a633c8fe6845675892e86ddfb7f764ed0c294c6d3642f8fd910c1e68fba08","last_reissued_at":"2026-07-05T06:27:45.375299Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:27:45.375299Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"EmbedDistill: A Geometric Knowledge Distillation for Information Retrieval","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Aditya Krishna Menon, Ankit Singh Rawat, Manzil Zaheer, Rob Fergus, Sadeep Jayasumana, Sanjiv Kumar, Seungyeon Kim, Veeranjaneyulu Sadhanala, Wittawat Jitkrittum","submitted_at":"2023-01-27T22:04:37Z","abstract_excerpt":"Large neural models (such as Transformers) achieve state-of-the-art performance for information retrieval (IR). In this paper, we aim to improve distillation methods that pave the way for the resource-efficient deployment of such models in practice. Inspired by our theoretical analysis of the teacher-student generalization gap for IR models, we propose a novel distillation approach that leverages the relative geometry among queries and documents learned by the large teacher model. Unlike existing teacher score-based distillation methods, our proposed approach employs embedding matching tasks t"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2301.12005","kind":"arxiv","version":2},"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/2301.12005/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":"2301.12005","created_at":"2026-07-05T06:27:45.375375+00:00"},{"alias_kind":"arxiv_version","alias_value":"2301.12005v2","created_at":"2026-07-05T06:27:45.375375+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2301.12005","created_at":"2026-07-05T06:27:45.375375+00:00"},{"alias_kind":"pith_short_12","alias_value":"ZGNGGPEP42CF","created_at":"2026-07-05T06:27:45.375375+00:00"},{"alias_kind":"pith_short_16","alias_value":"ZGNGGPEP42CFM5MJ","created_at":"2026-07-05T06:27:45.375375+00:00"},{"alias_kind":"pith_short_8","alias_value":"ZGNGGPEP","created_at":"2026-07-05T06:27:45.375375+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2602.15547","citing_title":"jina-embeddings-v5-text: Task-Targeted Embedding Distillation","ref_index":6,"is_internal_anchor":false},{"citing_arxiv_id":"2509.20354","citing_title":"EmbeddingGemma: Powerful and Lightweight Text Representations","ref_index":11,"is_internal_anchor":false},{"citing_arxiv_id":"2604.04734","citing_title":"Beyond Hard Negatives: The Importance of Score Distribution in Knowledge Distillation for Dense Retrieval","ref_index":13,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/ZGNGGPEP42CFM5MJF2DN3637OZ","json":"https://pith.science/pith/ZGNGGPEP42CFM5MJF2DN3637OZ.json","graph_json":"https://pith.science/api/pith-number/ZGNGGPEP42CFM5MJF2DN3637OZ/graph.json","events_json":"https://pith.science/api/pith-number/ZGNGGPEP42CFM5MJF2DN3637OZ/events.json","paper":"https://pith.science/paper/ZGNGGPEP"},"agent_actions":{"view_html":"https://pith.science/pith/ZGNGGPEP42CFM5MJF2DN3637OZ","download_json":"https://pith.science/pith/ZGNGGPEP42CFM5MJF2DN3637OZ.json","view_paper":"https://pith.science/paper/ZGNGGPEP","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2301.12005&json=true","fetch_graph":"https://pith.science/api/pith-number/ZGNGGPEP42CFM5MJF2DN3637OZ/graph.json","fetch_events":"https://pith.science/api/pith-number/ZGNGGPEP42CFM5MJF2DN3637OZ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/ZGNGGPEP42CFM5MJF2DN3637OZ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/ZGNGGPEP42CFM5MJF2DN3637OZ/action/storage_attestation","attest_author":"https://pith.science/pith/ZGNGGPEP42CFM5MJF2DN3637OZ/action/author_attestation","sign_citation":"https://pith.science/pith/ZGNGGPEP42CFM5MJF2DN3637OZ/action/citation_signature","submit_replication":"https://pith.science/pith/ZGNGGPEP42CFM5MJF2DN3637OZ/action/replication_record"}},"created_at":"2026-07-05T06:27:45.375375+00:00","updated_at":"2026-07-05T06:27:45.375375+00:00"}