{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:BNSYW7MRXBNSLOFDODW2DB4WZY","short_pith_number":"pith:BNSYW7MR","canonical_record":{"source":{"id":"2406.12335","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-06-18T07:01:11Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"fa48fdc71ae44a43f68e0501fa7ffa7f36099d0e4bcf3ffe7f14de13495769cd","abstract_canon_sha256":"33f54fb1ef3338ef280c6e12ea85bab39deba2ff3a84c986d5806b452b6aef01"},"schema_version":"1.0"},"canonical_sha256":"0b658b7d91b85b25b8a370eda18796ce1e1c8360163807ca636e8f73833b0edd","source":{"kind":"arxiv","id":"2406.12335","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2406.12335","created_at":"2026-07-05T09:14:34Z"},{"alias_kind":"arxiv_version","alias_value":"2406.12335v2","created_at":"2026-07-05T09:14:34Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2406.12335","created_at":"2026-07-05T09:14:34Z"},{"alias_kind":"pith_short_12","alias_value":"BNSYW7MRXBNS","created_at":"2026-07-05T09:14:34Z"},{"alias_kind":"pith_short_16","alias_value":"BNSYW7MRXBNSLOFD","created_at":"2026-07-05T09:14:34Z"},{"alias_kind":"pith_short_8","alias_value":"BNSYW7MR","created_at":"2026-07-05T09:14:34Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:BNSYW7MRXBNSLOFDODW2DB4WZY","target":"record","payload":{"canonical_record":{"source":{"id":"2406.12335","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-06-18T07:01:11Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"fa48fdc71ae44a43f68e0501fa7ffa7f36099d0e4bcf3ffe7f14de13495769cd","abstract_canon_sha256":"33f54fb1ef3338ef280c6e12ea85bab39deba2ff3a84c986d5806b452b6aef01"},"schema_version":"1.0"},"canonical_sha256":"0b658b7d91b85b25b8a370eda18796ce1e1c8360163807ca636e8f73833b0edd","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:14:34.684547Z","signature_b64":"IcYdqDg8cWQRPB2uuQU9vvwsexPwiNVcwnFkC0aNg89zLeocz0nL49g8wMTJSLBOpJ4FyftzF8jhTpWNl+q2Cg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"0b658b7d91b85b25b8a370eda18796ce1e1c8360163807ca636e8f73833b0edd","last_reissued_at":"2026-07-05T09:14:34.684013Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:14:34.684013Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2406.12335","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T09:14:34Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"tUcAGJHliKv0e289mfUTDiN3Nmqk42KZowAEwXAEL8Rtxopbxz+FuX49OLLMnG+rUWpTpakzcufwhdMGHVWxBg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-06T15:51:32.579930Z"},"content_sha256":"f0cf9990a59d4c7736b0276a8f21f8924c87180409fd214bea629dc2c7dc1ec2","schema_version":"1.0","event_id":"sha256:f0cf9990a59d4c7736b0276a8f21f8924c87180409fd214bea629dc2c7dc1ec2"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:BNSYW7MRXBNSLOFDODW2DB4WZY","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Attention Score is not All You Need for Token Importance Indicator in KV Cache Reduction: Value Also Matters","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Hidetaka Kamigaito, Taro Watanabe, Zhiyu Guo","submitted_at":"2024-06-18T07:01:11Z","abstract_excerpt":"Scaling the context size of large language models (LLMs) enables them to perform various new tasks, e.g., book summarization. However, the memory cost of the Key and Value (KV) cache in attention significantly limits the practical applications of LLMs. Recent works have explored token pruning for KV cache reduction in LLMs, relying solely on attention scores as a token importance indicator. However, our investigation into value vector norms revealed a notably non-uniform pattern questioning their reliance only on attention scores. Inspired by this, we propose a new method: Value-Aware Token Pr"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2406.12335","kind":"arxiv","version":2},"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/2406.12335/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":"2026-07-05T09:14:34Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"GBzVvOcQu0Wr7b9kn6LUGQDR6tpd2gmoMYyqNEe1NqLEKnuJz+bMMvAZMetQWAGslXIuEY0hui1oPZPBsSzfAw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-06T15:51:32.580278Z"},"content_sha256":"8b270ccbd28bb103328a2431af7f2edd58ac2980b2a267f570ab3204bf8457c3","schema_version":"1.0","event_id":"sha256:8b270ccbd28bb103328a2431af7f2edd58ac2980b2a267f570ab3204bf8457c3"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/BNSYW7MRXBNSLOFDODW2DB4WZY/bundle.json","state_url":"https://pith.science/pith/BNSYW7MRXBNSLOFDODW2DB4WZY/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/BNSYW7MRXBNSLOFDODW2DB4WZY/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":"2026-08-06T15:51:32Z","links":{"resolver":"https://pith.science/pith/BNSYW7MRXBNSLOFDODW2DB4WZY","bundle":"https://pith.science/pith/BNSYW7MRXBNSLOFDODW2DB4WZY/bundle.json","state":"https://pith.science/pith/BNSYW7MRXBNSLOFDODW2DB4WZY/state.json","well_known_bundle":"https://pith.science/.well-known/pith/BNSYW7MRXBNSLOFDODW2DB4WZY/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:BNSYW7MRXBNSLOFDODW2DB4WZY","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":"33f54fb1ef3338ef280c6e12ea85bab39deba2ff3a84c986d5806b452b6aef01","cross_cats_sorted":["cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-06-18T07:01:11Z","title_canon_sha256":"fa48fdc71ae44a43f68e0501fa7ffa7f36099d0e4bcf3ffe7f14de13495769cd"},"schema_version":"1.0","source":{"id":"2406.12335","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2406.12335","created_at":"2026-07-05T09:14:34Z"},{"alias_kind":"arxiv_version","alias_value":"2406.12335v2","created_at":"2026-07-05T09:14:34Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2406.12335","created_at":"2026-07-05T09:14:34Z"},{"alias_kind":"pith_short_12","alias_value":"BNSYW7MRXBNS","created_at":"2026-07-05T09:14:34Z"},{"alias_kind":"pith_short_16","alias_value":"BNSYW7MRXBNSLOFD","created_at":"2026-07-05T09:14:34Z"},{"alias_kind":"pith_short_8","alias_value":"BNSYW7MR","created_at":"2026-07-05T09:14:34Z"}],"graph_snapshots":[{"event_id":"sha256:8b270ccbd28bb103328a2431af7f2edd58ac2980b2a267f570ab3204bf8457c3","target":"graph","created_at":"2026-07-05T09:14:34Z","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/2406.12335/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Scaling the context size of large language models (LLMs) enables them to perform various new tasks, e.g., book summarization. However, the memory cost of the Key and Value (KV) cache in attention significantly limits the practical applications of LLMs. Recent works have explored token pruning for KV cache reduction in LLMs, relying solely on attention scores as a token importance indicator. However, our investigation into value vector norms revealed a notably non-uniform pattern questioning their reliance only on attention scores. Inspired by this, we propose a new method: Value-Aware Token Pr","authors_text":"Hidetaka Kamigaito, Taro Watanabe, Zhiyu Guo","cross_cats":["cs.LG"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-06-18T07:01:11Z","title":"Attention Score is not All You Need for Token Importance Indicator in KV Cache Reduction: Value Also Matters"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2406.12335","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:f0cf9990a59d4c7736b0276a8f21f8924c87180409fd214bea629dc2c7dc1ec2","target":"record","created_at":"2026-07-05T09:14:34Z","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":"33f54fb1ef3338ef280c6e12ea85bab39deba2ff3a84c986d5806b452b6aef01","cross_cats_sorted":["cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-06-18T07:01:11Z","title_canon_sha256":"fa48fdc71ae44a43f68e0501fa7ffa7f36099d0e4bcf3ffe7f14de13495769cd"},"schema_version":"1.0","source":{"id":"2406.12335","kind":"arxiv","version":2}},"canonical_sha256":"0b658b7d91b85b25b8a370eda18796ce1e1c8360163807ca636e8f73833b0edd","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"0b658b7d91b85b25b8a370eda18796ce1e1c8360163807ca636e8f73833b0edd","first_computed_at":"2026-07-05T09:14:34.684013Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T09:14:34.684013Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"IcYdqDg8cWQRPB2uuQU9vvwsexPwiNVcwnFkC0aNg89zLeocz0nL49g8wMTJSLBOpJ4FyftzF8jhTpWNl+q2Cg==","signature_status":"signed_v1","signed_at":"2026-07-05T09:14:34.684547Z","signed_message":"canonical_sha256_bytes"},"source_id":"2406.12335","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:f0cf9990a59d4c7736b0276a8f21f8924c87180409fd214bea629dc2c7dc1ec2","sha256:8b270ccbd28bb103328a2431af7f2edd58ac2980b2a267f570ab3204bf8457c3"],"state_sha256":"e7d8cf4c7aaf802827eb2d6f5bdbe969e06906e4c139d64bf0c00812fc0bd05e"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"XZsR4EXxH1mA8JUp+LAvq/oOminJmeD18Qq3RD7cygkvBTwE2LZHrRngE7Mzi4jcA1Rn6ttQRhB2Sefc+7ZGAQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-06T15:51:32.584414Z","bundle_sha256":"58ea6ea1ca5cb237415d8330aca47992499669fa47946960c6a652614437b800"}}