{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:2CCYNU7Q33VZF6YO2LNVYE3LO2","short_pith_number":"pith:2CCYNU7Q","schema_version":"1.0","canonical_sha256":"d08586d3f0deeb92fb0ed2db5c136b768d2028f51d057b69c1cdb9cd1611bc52","source":{"kind":"arxiv","id":"2405.19504","version":2},"attestation_state":"computed","paper":{"title":"MUVERA: Multi-Vector Retrieval via Fixed Dimensional Encodings","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.DB","cs.IR"],"primary_cat":"cs.DS","authors_text":"Jason Lee, Laxman Dhulipala, Majid Hadian, Rajesh Jayaram, Vahab Mirrokni","submitted_at":"2024-05-29T20:40:20Z","abstract_excerpt":"Neural embedding models have become a fundamental component of modern information retrieval (IR) pipelines. These models produce a single embedding $x \\in \\mathbb{R}^d$ per data-point, allowing for fast retrieval via highly optimized maximum inner product search (MIPS) algorithms. Recently, beginning with the landmark ColBERT paper, multi-vector models, which produce a set of embedding per data point, have achieved markedly superior performance for IR tasks. Unfortunately, using these models for IR is computationally expensive due to the increased complexity of multi-vector retrieval and scori"},"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":"2405.19504","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.DS","submitted_at":"2024-05-29T20:40:20Z","cross_cats_sorted":["cs.DB","cs.IR"],"title_canon_sha256":"d4598748b87d3a39b199016c49dd08b277fe8f570439763ccb6b3cce641cae0f","abstract_canon_sha256":"6c7857cd3c171951b2d3cd607ec5a3fe25556be374bbb21cfae5f67190da79fb"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-10T00:08:24.899203Z","signature_b64":"hmlSeIOXFBK9mgVas/CmIzng2RZUuEw5mW1FbeSG4SX8UFC3v3Un+aOyL8Nny91KUDlPUmEKSycCEUn+wexjDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"d08586d3f0deeb92fb0ed2db5c136b768d2028f51d057b69c1cdb9cd1611bc52","last_reissued_at":"2026-06-10T00:08:24.897954Z","signature_status":"signed_v1","first_computed_at":"2026-06-10T00:08:24.897954Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"MUVERA: Multi-Vector Retrieval via Fixed Dimensional Encodings","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.DB","cs.IR"],"primary_cat":"cs.DS","authors_text":"Jason Lee, Laxman Dhulipala, Majid Hadian, Rajesh Jayaram, Vahab Mirrokni","submitted_at":"2024-05-29T20:40:20Z","abstract_excerpt":"Neural embedding models have become a fundamental component of modern information retrieval (IR) pipelines. These models produce a single embedding $x \\in \\mathbb{R}^d$ per data-point, allowing for fast retrieval via highly optimized maximum inner product search (MIPS) algorithms. Recently, beginning with the landmark ColBERT paper, multi-vector models, which produce a set of embedding per data point, have achieved markedly superior performance for IR tasks. Unfortunately, using these models for IR is computationally expensive due to the increased complexity of multi-vector retrieval and scori"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2405.19504","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/2405.19504/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":"2405.19504","created_at":"2026-06-10T00:08:24.898134+00:00"},{"alias_kind":"arxiv_version","alias_value":"2405.19504v2","created_at":"2026-06-10T00:08:24.898134+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2405.19504","created_at":"2026-06-10T00:08:24.898134+00:00"},{"alias_kind":"pith_short_12","alias_value":"2CCYNU7Q33VZ","created_at":"2026-06-10T00:08:24.898134+00:00"},{"alias_kind":"pith_short_16","alias_value":"2CCYNU7Q33VZF6YO","created_at":"2026-06-10T00:08:24.898134+00:00"},{"alias_kind":"pith_short_8","alias_value":"2CCYNU7Q","created_at":"2026-06-10T00:08:24.898134+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2604.19566","citing_title":"Diagnosable ColBERT: Debugging Late-Interaction Retrieval Models Using a Learned Latent Space as Reference","ref_index":15,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/2CCYNU7Q33VZF6YO2LNVYE3LO2","json":"https://pith.science/pith/2CCYNU7Q33VZF6YO2LNVYE3LO2.json","graph_json":"https://pith.science/api/pith-number/2CCYNU7Q33VZF6YO2LNVYE3LO2/graph.json","events_json":"https://pith.science/api/pith-number/2CCYNU7Q33VZF6YO2LNVYE3LO2/events.json","paper":"https://pith.science/paper/2CCYNU7Q"},"agent_actions":{"view_html":"https://pith.science/pith/2CCYNU7Q33VZF6YO2LNVYE3LO2","download_json":"https://pith.science/pith/2CCYNU7Q33VZF6YO2LNVYE3LO2.json","view_paper":"https://pith.science/paper/2CCYNU7Q","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2405.19504&json=true","fetch_graph":"https://pith.science/api/pith-number/2CCYNU7Q33VZF6YO2LNVYE3LO2/graph.json","fetch_events":"https://pith.science/api/pith-number/2CCYNU7Q33VZF6YO2LNVYE3LO2/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/2CCYNU7Q33VZF6YO2LNVYE3LO2/action/timestamp_anchor","attest_storage":"https://pith.science/pith/2CCYNU7Q33VZF6YO2LNVYE3LO2/action/storage_attestation","attest_author":"https://pith.science/pith/2CCYNU7Q33VZF6YO2LNVYE3LO2/action/author_attestation","sign_citation":"https://pith.science/pith/2CCYNU7Q33VZF6YO2LNVYE3LO2/action/citation_signature","submit_replication":"https://pith.science/pith/2CCYNU7Q33VZF6YO2LNVYE3LO2/action/replication_record"}},"created_at":"2026-06-10T00:08:24.898134+00:00","updated_at":"2026-06-10T00:08:24.898134+00:00"}