{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:H7JXBLE24YJSILYC5CUIKXOG4V","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":"166332a4556fdc9209f844c3b0804d531d911317c3d432fe466f578aaa96f109","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-12-17T09:20:31Z","title_canon_sha256":"e16550f279c4965faf6f58351d827f3111b073c812d57c4d7824828ae55451ab"},"schema_version":"1.0","source":{"id":"2412.12706","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2412.12706","created_at":"2026-07-05T10:17:14Z"},{"alias_kind":"arxiv_version","alias_value":"2412.12706v2","created_at":"2026-07-05T10:17:14Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2412.12706","created_at":"2026-07-05T10:17:14Z"},{"alias_kind":"pith_short_12","alias_value":"H7JXBLE24YJS","created_at":"2026-07-05T10:17:14Z"},{"alias_kind":"pith_short_16","alias_value":"H7JXBLE24YJSILYC","created_at":"2026-07-05T10:17:14Z"},{"alias_kind":"pith_short_8","alias_value":"H7JXBLE2","created_at":"2026-07-05T10:17:14Z"}],"graph_snapshots":[{"event_id":"sha256:e0483cb459e7c1b45c131969cdbcb9495ec99a2dcaee1a1ab62ae7abbf00f52d","target":"graph","created_at":"2026-07-05T10:17:14Z","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/2412.12706/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"As large language models (LLMs) process increasing context windows, the memory usage of KV cache has become a critical bottleneck during inference. The mainstream KV compression methods, including KV pruning and KV quantization, primarily focus on either token or precision dimension separately. However, these works leaving the trade-off between these two orthogonal dimensions largely under-explored. In this paper, we comprehensively investigate the token-precision trade-off in KV cache compression.Experiments demonstrate that storing more tokens in the KV cache with lower precision,a strategy ","authors_text":"Chuqiao Kuang, Dawei Zhu, Jiebin Zhang, Lifeng Shang, Qun Liu, Sujian Li, Wenhao Wu, Xiaoguang Li, Yifan Song","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-12-17T09:20:31Z","title":"More Tokens, Lower Precision: Towards the Optimal Token-Precision Trade-off in KV Cache Compression"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2412.12706","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":"sha256:05b79a69042deeb76061ea64dc53d581c1a63f4958ade0c3a0de653de53a3454","target":"record","created_at":"2026-07-05T10:17:14Z","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":"166332a4556fdc9209f844c3b0804d531d911317c3d432fe466f578aaa96f109","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-12-17T09:20:31Z","title_canon_sha256":"e16550f279c4965faf6f58351d827f3111b073c812d57c4d7824828ae55451ab"},"schema_version":"1.0","source":{"id":"2412.12706","kind":"arxiv","version":2}},"canonical_sha256":"3fd370ac9ae613242f02e8a8855dc6e56ddbd096b4373b03cc0a093482c94c16","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"3fd370ac9ae613242f02e8a8855dc6e56ddbd096b4373b03cc0a093482c94c16","first_computed_at":"2026-07-05T10:17:14.697143Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T10:17:14.697143Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"wz155RDWXxQAjVN5c3zvjhzSie5pqIIO39os4ZkkpJQZaDDXCPazYdbW/EnuqMsXCL070Tytwbq7PS4gb2v3Cw==","signature_status":"signed_v1","signed_at":"2026-07-05T10:17:14.697681Z","signed_message":"canonical_sha256_bytes"},"source_id":"2412.12706","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:05b79a69042deeb76061ea64dc53d581c1a63f4958ade0c3a0de653de53a3454","sha256:e0483cb459e7c1b45c131969cdbcb9495ec99a2dcaee1a1ab62ae7abbf00f52d"],"state_sha256":"54a1af80d5f643db80430740d015cc1de8de605db50ee633f0a6d294a235decf"}