{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:OT6KG5FYO5ANRD3LHWAOSZ4SJR","short_pith_number":"pith:OT6KG5FY","canonical_record":{"source":{"id":"2506.15704","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-05-30T02:35:59Z","cross_cats_sorted":["cs.AI","cs.CL"],"title_canon_sha256":"5988fff61caa31a5b25ed15b88e8b0c1ebac9ab0c3f78d96ba8fc0a14c247110","abstract_canon_sha256":"53b2b0b42cda36e81b4651434fb5720e45cc29cf2f8da041ca11a01877591d23"},"schema_version":"1.0"},"canonical_sha256":"74fca374b87740d88f6b3d80e967924c4552ed109f6f0fdb6edb4152e3222201","source":{"kind":"arxiv","id":"2506.15704","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2506.15704","created_at":"2026-07-05T11:23:50Z"},{"alias_kind":"arxiv_version","alias_value":"2506.15704v1","created_at":"2026-07-05T11:23:50Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.15704","created_at":"2026-07-05T11:23:50Z"},{"alias_kind":"pith_short_12","alias_value":"OT6KG5FYO5AN","created_at":"2026-07-05T11:23:50Z"},{"alias_kind":"pith_short_16","alias_value":"OT6KG5FYO5ANRD3L","created_at":"2026-07-05T11:23:50Z"},{"alias_kind":"pith_short_8","alias_value":"OT6KG5FY","created_at":"2026-07-05T11:23:50Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:OT6KG5FYO5ANRD3LHWAOSZ4SJR","target":"record","payload":{"canonical_record":{"source":{"id":"2506.15704","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-05-30T02:35:59Z","cross_cats_sorted":["cs.AI","cs.CL"],"title_canon_sha256":"5988fff61caa31a5b25ed15b88e8b0c1ebac9ab0c3f78d96ba8fc0a14c247110","abstract_canon_sha256":"53b2b0b42cda36e81b4651434fb5720e45cc29cf2f8da041ca11a01877591d23"},"schema_version":"1.0"},"canonical_sha256":"74fca374b87740d88f6b3d80e967924c4552ed109f6f0fdb6edb4152e3222201","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:23:50.702965Z","signature_b64":"gJ6ZbttOBWIGPDjIwakwC9QIKekaMF6HG/d6jX4KRzEGd3O7WXTE1W8vRoAZbo21SLegH0j2ILFry2DO5l1RDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"74fca374b87740d88f6b3d80e967924c4552ed109f6f0fdb6edb4152e3222201","last_reissued_at":"2026-07-05T11:23:50.702458Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:23:50.702458Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2506.15704","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T11:23:50Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"TOg+pThjkE24WTvGdmHIdz4dLk2CYBIRcBLRTWhyBN+zoYm3Yftg0o+T1Jfsz1uKpSO54YboeFVqo50WImVRBA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-11T15:08:03.038267Z"},"content_sha256":"dd1ed8cf3161c4a849fd60c732d2534ce73a5defa8885eab19299310754f8e35","schema_version":"1.0","event_id":"sha256:dd1ed8cf3161c4a849fd60c732d2534ce73a5defa8885eab19299310754f8e35"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:OT6KG5FYO5ANRD3LHWAOSZ4SJR","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Learn from the Past: Fast Sparse Indexing for Large Language Model Decoding","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CL"],"primary_cat":"cs.LG","authors_text":"Feiyu Yao, Qian Wang","submitted_at":"2025-05-30T02:35:59Z","abstract_excerpt":"As large language models (LLMs) continue to support increasingly longer contexts, the memory demand for key-value (KV) caches during decoding grows rapidly, becoming a critical bottleneck in both GPU memory capacity and PCIe bandwidth. Sparse attention mechanisms alleviate this issue by computing attention weights only for selected key-value pairs. However, their indexing computation typically requires traversing all key vectors, resulting in significant computational and data transfer overhead. To reduce the cost of index retrieval, existing methods often treat each decoding step as an indepe"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.15704","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/2506.15704/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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T11:23:50Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"cQi9cgLYBX955nmtWDrfRN0msGZbrJQriwVzBpRoL12/4TKhttWtxVotKrCu+ic/m5NiARRGNVBC6+6X/ilsDw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-11T15:08:03.038828Z"},"content_sha256":"eb6730cda5995808cb9437d812581950fa1fe97edc0f3a3f43a0ccaf02db70af","schema_version":"1.0","event_id":"sha256:eb6730cda5995808cb9437d812581950fa1fe97edc0f3a3f43a0ccaf02db70af"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/OT6KG5FYO5ANRD3LHWAOSZ4SJR/bundle.json","state_url":"https://pith.science/pith/OT6KG5FYO5ANRD3LHWAOSZ4SJR/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/OT6KG5FYO5ANRD3LHWAOSZ4SJR/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-08-11T15:08:03Z","links":{"resolver":"https://pith.science/pith/OT6KG5FYO5ANRD3LHWAOSZ4SJR","bundle":"https://pith.science/pith/OT6KG5FYO5ANRD3LHWAOSZ4SJR/bundle.json","state":"https://pith.science/pith/OT6KG5FYO5ANRD3LHWAOSZ4SJR/state.json","well_known_bundle":"https://pith.science/.well-known/pith/OT6KG5FYO5ANRD3LHWAOSZ4SJR/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:OT6KG5FYO5ANRD3LHWAOSZ4SJR","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"53b2b0b42cda36e81b4651434fb5720e45cc29cf2f8da041ca11a01877591d23","cross_cats_sorted":["cs.AI","cs.CL"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-05-30T02:35:59Z","title_canon_sha256":"5988fff61caa31a5b25ed15b88e8b0c1ebac9ab0c3f78d96ba8fc0a14c247110"},"schema_version":"1.0","source":{"id":"2506.15704","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2506.15704","created_at":"2026-07-05T11:23:50Z"},{"alias_kind":"arxiv_version","alias_value":"2506.15704v1","created_at":"2026-07-05T11:23:50Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.15704","created_at":"2026-07-05T11:23:50Z"},{"alias_kind":"pith_short_12","alias_value":"OT6KG5FYO5AN","created_at":"2026-07-05T11:23:50Z"},{"alias_kind":"pith_short_16","alias_value":"OT6KG5FYO5ANRD3L","created_at":"2026-07-05T11:23:50Z"},{"alias_kind":"pith_short_8","alias_value":"OT6KG5FY","created_at":"2026-07-05T11:23:50Z"}],"graph_snapshots":[{"event_id":"sha256:eb6730cda5995808cb9437d812581950fa1fe97edc0f3a3f43a0ccaf02db70af","target":"graph","created_at":"2026-07-05T11:23:50Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2506.15704/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"As large language models (LLMs) continue to support increasingly longer contexts, the memory demand for key-value (KV) caches during decoding grows rapidly, becoming a critical bottleneck in both GPU memory capacity and PCIe bandwidth. Sparse attention mechanisms alleviate this issue by computing attention weights only for selected key-value pairs. However, their indexing computation typically requires traversing all key vectors, resulting in significant computational and data transfer overhead. To reduce the cost of index retrieval, existing methods often treat each decoding step as an indepe","authors_text":"Feiyu Yao, Qian Wang","cross_cats":["cs.AI","cs.CL"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-05-30T02:35:59Z","title":"Learn from the Past: Fast Sparse Indexing for Large Language Model Decoding"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.15704","kind":"arxiv","version":1},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:dd1ed8cf3161c4a849fd60c732d2534ce73a5defa8885eab19299310754f8e35","target":"record","created_at":"2026-07-05T11:23:50Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"53b2b0b42cda36e81b4651434fb5720e45cc29cf2f8da041ca11a01877591d23","cross_cats_sorted":["cs.AI","cs.CL"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-05-30T02:35:59Z","title_canon_sha256":"5988fff61caa31a5b25ed15b88e8b0c1ebac9ab0c3f78d96ba8fc0a14c247110"},"schema_version":"1.0","source":{"id":"2506.15704","kind":"arxiv","version":1}},"canonical_sha256":"74fca374b87740d88f6b3d80e967924c4552ed109f6f0fdb6edb4152e3222201","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"74fca374b87740d88f6b3d80e967924c4552ed109f6f0fdb6edb4152e3222201","first_computed_at":"2026-07-05T11:23:50.702458Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:23:50.702458Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"gJ6ZbttOBWIGPDjIwakwC9QIKekaMF6HG/d6jX4KRzEGd3O7WXTE1W8vRoAZbo21SLegH0j2ILFry2DO5l1RDg==","signature_status":"signed_v1","signed_at":"2026-07-05T11:23:50.702965Z","signed_message":"canonical_sha256_bytes"},"source_id":"2506.15704","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:dd1ed8cf3161c4a849fd60c732d2534ce73a5defa8885eab19299310754f8e35","sha256:eb6730cda5995808cb9437d812581950fa1fe97edc0f3a3f43a0ccaf02db70af"],"state_sha256":"53094b44dd9cd4c3c1642e069c920a5b4348eee32d6a5a5cc84c9e234a5ff273"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"A1raCKfwwobC2dbP0oVmxKsbnLAf2Ug+Z61xaVsZHjahBfegLhlsBNpNNstsDYUNIaSVc7ZTpB/RTYgHzpDiBA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-11T15:08:03.045904Z","bundle_sha256":"ee88a52912cc5a8c39de8e8e8976839c52885825bef03700174cd939b606b153"}}