{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:VZ5VDBWXKCEBIESYM3GA3B4UJN","short_pith_number":"pith:VZ5VDBWX","schema_version":"1.0","canonical_sha256":"ae7b5186d7508814125866cc0d87944b5772b19dd8e22e8d8f35d8db8525d4d0","source":{"kind":"arxiv","id":"2605.29517","version":1},"attestation_state":"computed","paper":{"title":"FLASH-MAXSIM: IO-Aware Fused Kernels for Late-Interaction Scoring","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.IR","authors_text":"Adi Raz Goldfarb, Daniel Ezer, Idan Friedman, Roi pony, Udi Barzelay","submitted_at":"2026-05-28T07:38:27Z","abstract_excerpt":"Late-interaction retrieval (ColBERT, ColPali) scores a query against a document with the MaxSim operator: for every query token, the maximum similarity over the document tokens, summed over\n  query tokens. The standard implementation materializes the full query-token x document-token similarity tensor in GPU memory; for visual ColPali at 10K documents this tensor alone is 21 GB in\n  FP16, created only to be reduced to one score per document and discarded. It exhausts a 40 GB GPU and bounds the achievable batch size in both inference and training. We present\n  Flash-MaxSim, an IO-aware fused GP"},"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":"2605.29517","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.IR","submitted_at":"2026-05-28T07:38:27Z","cross_cats_sorted":[],"title_canon_sha256":"1e8b4aa58d3ae119c7fea0965ec7c16a6bc25def84edd715154ba4edf20b1bda","abstract_canon_sha256":"fbd483db36a83cb606c66cac7ab009a9b229487a1cb115222db1828ecb472a64"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-29T01:05:44.544168Z","signature_b64":"pV1jq+ti9nE2mt3CGBbPTVkzg1n9WBQGv/TKC4yYIRHX/oA2HI/gMHxpjTl2UazaxT/QcGTDXeh9nk7UJmW/CA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"ae7b5186d7508814125866cc0d87944b5772b19dd8e22e8d8f35d8db8525d4d0","last_reissued_at":"2026-05-29T01:05:44.543423Z","signature_status":"signed_v1","first_computed_at":"2026-05-29T01:05:44.543423Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"FLASH-MAXSIM: IO-Aware Fused Kernels for Late-Interaction Scoring","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.IR","authors_text":"Adi Raz Goldfarb, Daniel Ezer, Idan Friedman, Roi pony, Udi Barzelay","submitted_at":"2026-05-28T07:38:27Z","abstract_excerpt":"Late-interaction retrieval (ColBERT, ColPali) scores a query against a document with the MaxSim operator: for every query token, the maximum similarity over the document tokens, summed over\n  query tokens. The standard implementation materializes the full query-token x document-token similarity tensor in GPU memory; for visual ColPali at 10K documents this tensor alone is 21 GB in\n  FP16, created only to be reduced to one score per document and discarded. It exhausts a 40 GB GPU and bounds the achievable batch size in both inference and training. We present\n  Flash-MaxSim, an IO-aware fused GP"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.29517","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/2605.29517/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":"2605.29517","created_at":"2026-05-29T01:05:44.543546+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.29517v1","created_at":"2026-05-29T01:05:44.543546+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.29517","created_at":"2026-05-29T01:05:44.543546+00:00"},{"alias_kind":"pith_short_12","alias_value":"VZ5VDBWXKCEB","created_at":"2026-05-29T01:05:44.543546+00:00"},{"alias_kind":"pith_short_16","alias_value":"VZ5VDBWXKCEBIESY","created_at":"2026-05-29T01:05:44.543546+00:00"},{"alias_kind":"pith_short_8","alias_value":"VZ5VDBWX","created_at":"2026-05-29T01:05:44.543546+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/VZ5VDBWXKCEBIESYM3GA3B4UJN","json":"https://pith.science/pith/VZ5VDBWXKCEBIESYM3GA3B4UJN.json","graph_json":"https://pith.science/api/pith-number/VZ5VDBWXKCEBIESYM3GA3B4UJN/graph.json","events_json":"https://pith.science/api/pith-number/VZ5VDBWXKCEBIESYM3GA3B4UJN/events.json","paper":"https://pith.science/paper/VZ5VDBWX"},"agent_actions":{"view_html":"https://pith.science/pith/VZ5VDBWXKCEBIESYM3GA3B4UJN","download_json":"https://pith.science/pith/VZ5VDBWXKCEBIESYM3GA3B4UJN.json","view_paper":"https://pith.science/paper/VZ5VDBWX","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.29517&json=true","fetch_graph":"https://pith.science/api/pith-number/VZ5VDBWXKCEBIESYM3GA3B4UJN/graph.json","fetch_events":"https://pith.science/api/pith-number/VZ5VDBWXKCEBIESYM3GA3B4UJN/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/VZ5VDBWXKCEBIESYM3GA3B4UJN/action/timestamp_anchor","attest_storage":"https://pith.science/pith/VZ5VDBWXKCEBIESYM3GA3B4UJN/action/storage_attestation","attest_author":"https://pith.science/pith/VZ5VDBWXKCEBIESYM3GA3B4UJN/action/author_attestation","sign_citation":"https://pith.science/pith/VZ5VDBWXKCEBIESYM3GA3B4UJN/action/citation_signature","submit_replication":"https://pith.science/pith/VZ5VDBWXKCEBIESYM3GA3B4UJN/action/replication_record"}},"created_at":"2026-05-29T01:05:44.543546+00:00","updated_at":"2026-05-29T01:05:44.543546+00:00"}