{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:R3X6BOBDL2TLAZDPCTXJVNKMZY","short_pith_number":"pith:R3X6BOBD","schema_version":"1.0","canonical_sha256":"8eefe0b8235ea6b0646f14ee9ab54cce00aca9865304b0740959411bb1a67158","source":{"kind":"arxiv","id":"2602.02827","version":2},"attestation_state":"computed","paper":{"title":"Col-Bandit: Query-Time Top-$K$ Estimation for Late-Interaction Retrieval","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.IR","authors_text":"Adi Raz Goldfarb, Eli Schwartz, Idan Friedman, Oshri Naparstek, Roi pony, Udi Barzelay","submitted_at":"2026-02-02T21:27:01Z","abstract_excerpt":"Multi-vector late-interaction retrievers such as ColBERT achieve state-of-the-art quality, but their query-time cost is dominated by exhaustively computing token-level MaxSim interactions for every candidate document. The MaxSim scores of $N$ candidates against $T$ query tokens form an $N\\times T$ matrix whose row-sums are the late-interaction scores, and identifying the top-$K$ rarely requires every entry. We introduce Col-Bandit, a query-time estimator of the exhaustive-MaxSim top-$K$: it reveals matrix entries in batches, maintains a finite-population Bernstein-Serfling confidence interval "},"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":"2602.02827","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.IR","submitted_at":"2026-02-02T21:27:01Z","cross_cats_sorted":[],"title_canon_sha256":"d173709172b5d50a3d9ae799aa27f154573f6dab7797b67f524e95c8144e03a0","abstract_canon_sha256":"4d68681543d6fe4e3988eb4152a1877249e5e30164622cbbad8706cc365f5b7c"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-03T02:05:45.388183Z","signature_b64":"ELtKymfWpEmlNnET8QvHFzyWEOpEghvL96gKcn9vb86V625q3aWkyuBtwP4amrhJUguZ2tl9zkZavxLVeSZ1Dw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"8eefe0b8235ea6b0646f14ee9ab54cce00aca9865304b0740959411bb1a67158","last_reissued_at":"2026-06-03T02:05:45.387669Z","signature_status":"signed_v1","first_computed_at":"2026-06-03T02:05:45.387669Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Col-Bandit: Query-Time Top-$K$ Estimation for Late-Interaction Retrieval","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.IR","authors_text":"Adi Raz Goldfarb, Eli Schwartz, Idan Friedman, Oshri Naparstek, Roi pony, Udi Barzelay","submitted_at":"2026-02-02T21:27:01Z","abstract_excerpt":"Multi-vector late-interaction retrievers such as ColBERT achieve state-of-the-art quality, but their query-time cost is dominated by exhaustively computing token-level MaxSim interactions for every candidate document. The MaxSim scores of $N$ candidates against $T$ query tokens form an $N\\times T$ matrix whose row-sums are the late-interaction scores, and identifying the top-$K$ rarely requires every entry. We introduce Col-Bandit, a query-time estimator of the exhaustive-MaxSim top-$K$: it reveals matrix entries in batches, maintains a finite-population Bernstein-Serfling confidence interval "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2602.02827","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/2602.02827/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":"2602.02827","created_at":"2026-06-03T02:05:45.387732+00:00"},{"alias_kind":"arxiv_version","alias_value":"2602.02827v2","created_at":"2026-06-03T02:05:45.387732+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2602.02827","created_at":"2026-06-03T02:05:45.387732+00:00"},{"alias_kind":"pith_short_12","alias_value":"R3X6BOBDL2TL","created_at":"2026-06-03T02:05:45.387732+00:00"},{"alias_kind":"pith_short_16","alias_value":"R3X6BOBDL2TLAZDP","created_at":"2026-06-03T02:05:45.387732+00:00"},{"alias_kind":"pith_short_8","alias_value":"R3X6BOBD","created_at":"2026-06-03T02:05:45.387732+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/R3X6BOBDL2TLAZDPCTXJVNKMZY","json":"https://pith.science/pith/R3X6BOBDL2TLAZDPCTXJVNKMZY.json","graph_json":"https://pith.science/api/pith-number/R3X6BOBDL2TLAZDPCTXJVNKMZY/graph.json","events_json":"https://pith.science/api/pith-number/R3X6BOBDL2TLAZDPCTXJVNKMZY/events.json","paper":"https://pith.science/paper/R3X6BOBD"},"agent_actions":{"view_html":"https://pith.science/pith/R3X6BOBDL2TLAZDPCTXJVNKMZY","download_json":"https://pith.science/pith/R3X6BOBDL2TLAZDPCTXJVNKMZY.json","view_paper":"https://pith.science/paper/R3X6BOBD","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2602.02827&json=true","fetch_graph":"https://pith.science/api/pith-number/R3X6BOBDL2TLAZDPCTXJVNKMZY/graph.json","fetch_events":"https://pith.science/api/pith-number/R3X6BOBDL2TLAZDPCTXJVNKMZY/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/R3X6BOBDL2TLAZDPCTXJVNKMZY/action/timestamp_anchor","attest_storage":"https://pith.science/pith/R3X6BOBDL2TLAZDPCTXJVNKMZY/action/storage_attestation","attest_author":"https://pith.science/pith/R3X6BOBDL2TLAZDPCTXJVNKMZY/action/author_attestation","sign_citation":"https://pith.science/pith/R3X6BOBDL2TLAZDPCTXJVNKMZY/action/citation_signature","submit_replication":"https://pith.science/pith/R3X6BOBDL2TLAZDPCTXJVNKMZY/action/replication_record"}},"created_at":"2026-06-03T02:05:45.387732+00:00","updated_at":"2026-06-03T02:05:45.387732+00:00"}