{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:I7FRUQEQRFHZE4DXMO7FAXONRY","short_pith_number":"pith:I7FRUQEQ","schema_version":"1.0","canonical_sha256":"47cb1a4090894f92707763be505dcd8e0b5c4e1c51439d216cfb3ddb64ed085a","source":{"kind":"arxiv","id":"2607.27692","version":1},"attestation_state":"computed","paper":{"title":"Recall Before You Rank: Similarity-Guided Top-$K$ Reuse for Efficient Long-Context Attention","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Hanyong Shao, Kechao Tang, Ru Huang, Wenshuai Yao, Wenyong Zhou, Yizhe Chen, Yuannuo Feng, Zhiyuan Ning","submitted_at":"2026-07-30T05:25:23Z","abstract_excerpt":"Top-$K$ sparse attention reduces the cost of Softmax and value aggregation by attending to only a small subset of key--value (KV) entries. However, identifying this subset still requires scoring the current query against the full KV cache and performing global Top-$K$ selection, leaving selector cost linear in context length and limiting the practical efficiency of sparse attention for long-context decoding. In this paper, we introduce ReTopK, a training-free method that accelerates dynamic Top-$K$ attention by reusing historical retrieval decisions. ReTopK builds on the observation that simil"},"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":"2607.27692","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2026-07-30T05:25:23Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"cce521f1fb3763583d2bb5b84503f05ef69ba25fa9f59deb708a83fd65b86f9d","abstract_canon_sha256":"1f351767b35c548f18e738307285fc9349894f89bd52b6a8047b73676654c61c"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"47cb1a4090894f92707763be505dcd8e0b5c4e1c51439d216cfb3ddb64ed085a","last_reissued_at":"2026-07-31T01:29:44.891787Z","signature_status":"unsigned_v0","first_computed_at":"2026-07-31T01:29:44.891787Z"},"graph_snapshot":{"paper":{"title":"Recall Before You Rank: Similarity-Guided Top-$K$ Reuse for Efficient Long-Context Attention","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Hanyong Shao, Kechao Tang, Ru Huang, Wenshuai Yao, Wenyong Zhou, Yizhe Chen, Yuannuo Feng, Zhiyuan Ning","submitted_at":"2026-07-30T05:25:23Z","abstract_excerpt":"Top-$K$ sparse attention reduces the cost of Softmax and value aggregation by attending to only a small subset of key--value (KV) entries. However, identifying this subset still requires scoring the current query against the full KV cache and performing global Top-$K$ selection, leaving selector cost linear in context length and limiting the practical efficiency of sparse attention for long-context decoding. In this paper, we introduce ReTopK, a training-free method that accelerates dynamic Top-$K$ attention by reusing historical retrieval decisions. ReTopK builds on the observation that simil"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.27692","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/2607.27692/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":"2607.27692","created_at":"2026-07-31T01:29:44.895015+00:00"},{"alias_kind":"arxiv_version","alias_value":"2607.27692v1","created_at":"2026-07-31T01:29:44.895015+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.27692","created_at":"2026-07-31T01:29:44.895015+00:00"},{"alias_kind":"pith_short_12","alias_value":"I7FRUQEQRFHZ","created_at":"2026-07-31T01:29:44.895015+00:00"},{"alias_kind":"pith_short_16","alias_value":"I7FRUQEQRFHZE4DX","created_at":"2026-07-31T01:29:44.895015+00:00"},{"alias_kind":"pith_short_8","alias_value":"I7FRUQEQ","created_at":"2026-07-31T01:29:44.895015+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/I7FRUQEQRFHZE4DXMO7FAXONRY","json":"https://pith.science/pith/I7FRUQEQRFHZE4DXMO7FAXONRY.json","graph_json":"https://pith.science/api/pith-number/I7FRUQEQRFHZE4DXMO7FAXONRY/graph.json","events_json":"https://pith.science/api/pith-number/I7FRUQEQRFHZE4DXMO7FAXONRY/events.json","paper":"https://pith.science/paper/I7FRUQEQ"},"agent_actions":{"view_html":"https://pith.science/pith/I7FRUQEQRFHZE4DXMO7FAXONRY","download_json":"https://pith.science/pith/I7FRUQEQRFHZE4DXMO7FAXONRY.json","view_paper":"https://pith.science/paper/I7FRUQEQ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2607.27692&json=true","fetch_graph":"https://pith.science/api/pith-number/I7FRUQEQRFHZE4DXMO7FAXONRY/graph.json","fetch_events":"https://pith.science/api/pith-number/I7FRUQEQRFHZE4DXMO7FAXONRY/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/I7FRUQEQRFHZE4DXMO7FAXONRY/action/timestamp_anchor","attest_storage":"https://pith.science/pith/I7FRUQEQRFHZE4DXMO7FAXONRY/action/storage_attestation","attest_author":"https://pith.science/pith/I7FRUQEQRFHZE4DXMO7FAXONRY/action/author_attestation","sign_citation":"https://pith.science/pith/I7FRUQEQRFHZE4DXMO7FAXONRY/action/citation_signature","submit_replication":"https://pith.science/pith/I7FRUQEQRFHZE4DXMO7FAXONRY/action/replication_record"}},"created_at":"2026-07-31T01:29:44.895015+00:00","updated_at":"2026-07-31T01:29:44.895015+00:00"}