{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:NTXQ4WRSI7QSMCUID4ABUWFBOR","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":"484dfc167d0eba554afb19292b25cf9b65f8bf595db4e42c1aa4f458d6f6dba5","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-02-18T09:11:51Z","title_canon_sha256":"3eb6dee532883181a3554dca9e2ddf4ab6297b9330ee216224e67a01b8619eb1"},"schema_version":"1.0","source":{"id":"2502.12665","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2502.12665","created_at":"2026-07-05T11:14:53Z"},{"alias_kind":"arxiv_version","alias_value":"2502.12665v2","created_at":"2026-07-05T11:14:53Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.12665","created_at":"2026-07-05T11:14:53Z"},{"alias_kind":"pith_short_12","alias_value":"NTXQ4WRSI7QS","created_at":"2026-07-05T11:14:53Z"},{"alias_kind":"pith_short_16","alias_value":"NTXQ4WRSI7QSMCUI","created_at":"2026-07-05T11:14:53Z"},{"alias_kind":"pith_short_8","alias_value":"NTXQ4WRS","created_at":"2026-07-05T11:14:53Z"}],"graph_snapshots":[{"event_id":"sha256:0269348c07209731a812b975bc362a8a3017b1c5755897041e0b184ae9185907","target":"graph","created_at":"2026-07-05T11:14:53Z","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/2502.12665/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Long context large language models (LLMs) pose significant challenges for efficient serving due to the large memory footprint and high access overhead of KV cache. Retrieval-based KV cache reduction methods can mitigate these challenges, typically by offloading the complete KV cache to CPU and retrieving necessary tokens on demand during inference. However, these methods still suffer from unsatisfactory accuracy degradation and extra retrieval overhead. To address these limitations, this paper proposes A$^2$ATS, a novel retrieval-based KV cache reduction method. A$^2$ATS aims to obtain an accu","authors_text":"Chun Jason Xue, Junhui He, Junna Xing, Nan Wang, Peng Zhou, Qiang Liu, Qingan Li, Rui Xu, Shangyu Wu","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-02-18T09:11:51Z","title":"A$^2$ATS: Retrieval-Based KV Cache Reduction via Windowed Rotary Position Embedding and Query-Aware Vector Quantization"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.12665","kind":"arxiv","version":2},"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":"sha256:1d1a14cc579e7a9daee14ebce8621e98c133a5b2bd5303b67e3a160ab7903334","target":"record","created_at":"2026-07-05T11:14:53Z","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":"484dfc167d0eba554afb19292b25cf9b65f8bf595db4e42c1aa4f458d6f6dba5","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-02-18T09:11:51Z","title_canon_sha256":"3eb6dee532883181a3554dca9e2ddf4ab6297b9330ee216224e67a01b8619eb1"},"schema_version":"1.0","source":{"id":"2502.12665","kind":"arxiv","version":2}},"canonical_sha256":"6cef0e5a3247e1260a881f001a58a174508fa01619d131cd714d53b9f5a0b827","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"6cef0e5a3247e1260a881f001a58a174508fa01619d131cd714d53b9f5a0b827","first_computed_at":"2026-07-05T11:14:53.960530Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:14:53.960530Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"Vi6Wn6WMs2VGRtbEU/Gk5LN4MPiw1WrgBKZXgfWHcSYUdhqFtnJ4Qho6oCmTHjJsebC3No6aLQe3AcJkz1H4CA==","signature_status":"signed_v1","signed_at":"2026-07-05T11:14:53.960998Z","signed_message":"canonical_sha256_bytes"},"source_id":"2502.12665","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:1d1a14cc579e7a9daee14ebce8621e98c133a5b2bd5303b67e3a160ab7903334","sha256:0269348c07209731a812b975bc362a8a3017b1c5755897041e0b184ae9185907"],"state_sha256":"227d1a9ba11c1a9897d358267f0e042c75443f2b7592377dca307523c25812bf"}