{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:5TXPIUIPS7JPFLUMIJVH5U7XAU","short_pith_number":"pith:5TXPIUIP","canonical_record":{"source":{"id":"2508.02401","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2025-08-04T13:26:16Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"ae8791caaf8f95f5d3d4cf605dbe759d134e74a72594aa0a75203133d789dc35","abstract_canon_sha256":"df0a68b5b7327d643f36814c5391a7bf336ea7f12736e190632fe3033995ac38"},"schema_version":"1.0"},"canonical_sha256":"eceef4510f97d2f2ae8c426a7ed3f70537ac8abda41c5e401f3d5747f79c7788","source":{"kind":"arxiv","id":"2508.02401","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2508.02401","created_at":"2026-07-05T11:48:13Z"},{"alias_kind":"arxiv_version","alias_value":"2508.02401v1","created_at":"2026-07-05T11:48:13Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2508.02401","created_at":"2026-07-05T11:48:13Z"},{"alias_kind":"pith_short_12","alias_value":"5TXPIUIPS7JP","created_at":"2026-07-05T11:48:13Z"},{"alias_kind":"pith_short_16","alias_value":"5TXPIUIPS7JPFLUM","created_at":"2026-07-05T11:48:13Z"},{"alias_kind":"pith_short_8","alias_value":"5TXPIUIP","created_at":"2026-07-05T11:48:13Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:5TXPIUIPS7JPFLUMIJVH5U7XAU","target":"record","payload":{"canonical_record":{"source":{"id":"2508.02401","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2025-08-04T13:26:16Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"ae8791caaf8f95f5d3d4cf605dbe759d134e74a72594aa0a75203133d789dc35","abstract_canon_sha256":"df0a68b5b7327d643f36814c5391a7bf336ea7f12736e190632fe3033995ac38"},"schema_version":"1.0"},"canonical_sha256":"eceef4510f97d2f2ae8c426a7ed3f70537ac8abda41c5e401f3d5747f79c7788","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:48:13.788029Z","signature_b64":"ohdW+VbNZjHYUFPdN4a9HdZuLRL/6GpJ13oJHNcO2mwwXgK8Vm6Bj5XyZP6HNtZF5Fsy02ntZ3r+obq0IK3QAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"eceef4510f97d2f2ae8c426a7ed3f70537ac8abda41c5e401f3d5747f79c7788","last_reissued_at":"2026-07-05T11:48:13.787475Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:48:13.787475Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2508.02401","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":"2026-07-05T11:48:13Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"7++ptBI/Z7fhXexBa69a6vPXWaiqQPUFbUeFj3WaOf5fHV2FNKPZd+Utj5w4yIQOtfXvD77sJKa8YAKgyMphCg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-06T22:52:09.972448Z"},"content_sha256":"5780852aae47c6b76216958ef11d20ca04603b3988187afbf9b97ed5a098cc40","schema_version":"1.0","event_id":"sha256:5780852aae47c6b76216958ef11d20ca04603b3988187afbf9b97ed5a098cc40"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:5TXPIUIPS7JPFLUMIJVH5U7XAU","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"CompressKV: Semantic Retrieval Heads Know What Tokens are Not Important Before Generation","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Bing Li, Grace Li Zhang, Jingcun Wang, Olga Kondrateva, Xiaolin Lin, Yiyu Shi","submitted_at":"2025-08-04T13:26:16Z","abstract_excerpt":"Recent advances in large language models (LLMs) have significantly boosted long-context processing. However, the increasing key-value (KV) cache size poses critical challenges to memory and execution efficiency. Most KV cache compression methods rely on heuristic token eviction using all attention heads in Grouped Query Attention (GQA)-based LLMs. This method ignores the different functionalities of attention heads, leading to the eviction of critical tokens and thus degrades the performance of LLMs.\n  To address the issue above, instead of using all the attention heads in GQA-based LLMs to de"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2508.02401","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/2508.02401/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":"2026-07-05T11:48:13Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"V4KNCHtrHjKgHRWqY6NVvWkz4ojxa0DhmWrnhqH8wndr5Ug/BAbTuxeOKA0ik8O2O5I0N4bA2rvWCinuLgVjAQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-06T22:52:09.972956Z"},"content_sha256":"e4c50440972332f53e2c6b347493b4e9d96a5fe26b552fcd8c4dd9230c3a2dbe","schema_version":"1.0","event_id":"sha256:e4c50440972332f53e2c6b347493b4e9d96a5fe26b552fcd8c4dd9230c3a2dbe"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/5TXPIUIPS7JPFLUMIJVH5U7XAU/bundle.json","state_url":"https://pith.science/pith/5TXPIUIPS7JPFLUMIJVH5U7XAU/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/5TXPIUIPS7JPFLUMIJVH5U7XAU/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":"2026-08-06T22:52:09Z","links":{"resolver":"https://pith.science/pith/5TXPIUIPS7JPFLUMIJVH5U7XAU","bundle":"https://pith.science/pith/5TXPIUIPS7JPFLUMIJVH5U7XAU/bundle.json","state":"https://pith.science/pith/5TXPIUIPS7JPFLUMIJVH5U7XAU/state.json","well_known_bundle":"https://pith.science/.well-known/pith/5TXPIUIPS7JPFLUMIJVH5U7XAU/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:5TXPIUIPS7JPFLUMIJVH5U7XAU","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":"df0a68b5b7327d643f36814c5391a7bf336ea7f12736e190632fe3033995ac38","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2025-08-04T13:26:16Z","title_canon_sha256":"ae8791caaf8f95f5d3d4cf605dbe759d134e74a72594aa0a75203133d789dc35"},"schema_version":"1.0","source":{"id":"2508.02401","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2508.02401","created_at":"2026-07-05T11:48:13Z"},{"alias_kind":"arxiv_version","alias_value":"2508.02401v1","created_at":"2026-07-05T11:48:13Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2508.02401","created_at":"2026-07-05T11:48:13Z"},{"alias_kind":"pith_short_12","alias_value":"5TXPIUIPS7JP","created_at":"2026-07-05T11:48:13Z"},{"alias_kind":"pith_short_16","alias_value":"5TXPIUIPS7JPFLUM","created_at":"2026-07-05T11:48:13Z"},{"alias_kind":"pith_short_8","alias_value":"5TXPIUIP","created_at":"2026-07-05T11:48:13Z"}],"graph_snapshots":[{"event_id":"sha256:e4c50440972332f53e2c6b347493b4e9d96a5fe26b552fcd8c4dd9230c3a2dbe","target":"graph","created_at":"2026-07-05T11:48:13Z","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/2508.02401/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Recent advances in large language models (LLMs) have significantly boosted long-context processing. However, the increasing key-value (KV) cache size poses critical challenges to memory and execution efficiency. Most KV cache compression methods rely on heuristic token eviction using all attention heads in Grouped Query Attention (GQA)-based LLMs. This method ignores the different functionalities of attention heads, leading to the eviction of critical tokens and thus degrades the performance of LLMs.\n  To address the issue above, instead of using all the attention heads in GQA-based LLMs to de","authors_text":"Bing Li, Grace Li Zhang, Jingcun Wang, Olga Kondrateva, Xiaolin Lin, Yiyu Shi","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2025-08-04T13:26:16Z","title":"CompressKV: Semantic Retrieval Heads Know What Tokens are Not Important Before Generation"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2508.02401","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":"sha256:5780852aae47c6b76216958ef11d20ca04603b3988187afbf9b97ed5a098cc40","target":"record","created_at":"2026-07-05T11:48:13Z","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":"df0a68b5b7327d643f36814c5391a7bf336ea7f12736e190632fe3033995ac38","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2025-08-04T13:26:16Z","title_canon_sha256":"ae8791caaf8f95f5d3d4cf605dbe759d134e74a72594aa0a75203133d789dc35"},"schema_version":"1.0","source":{"id":"2508.02401","kind":"arxiv","version":1}},"canonical_sha256":"eceef4510f97d2f2ae8c426a7ed3f70537ac8abda41c5e401f3d5747f79c7788","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"eceef4510f97d2f2ae8c426a7ed3f70537ac8abda41c5e401f3d5747f79c7788","first_computed_at":"2026-07-05T11:48:13.787475Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:48:13.787475Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"ohdW+VbNZjHYUFPdN4a9HdZuLRL/6GpJ13oJHNcO2mwwXgK8Vm6Bj5XyZP6HNtZF5Fsy02ntZ3r+obq0IK3QAg==","signature_status":"signed_v1","signed_at":"2026-07-05T11:48:13.788029Z","signed_message":"canonical_sha256_bytes"},"source_id":"2508.02401","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:5780852aae47c6b76216958ef11d20ca04603b3988187afbf9b97ed5a098cc40","sha256:e4c50440972332f53e2c6b347493b4e9d96a5fe26b552fcd8c4dd9230c3a2dbe"],"state_sha256":"e5b2c6cf734f05350c4143e9ded5144a6bbb3139daf3807698bcb5447a539558"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"OhZRRIMCPOWeU+iQFOJez9Y75X6Tnr0V92QkJJJQkw0/YjsZUvNanNmc3gkXQOsxxOYfEJLXoaiu8QTNwj9BBA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-06T22:52:09.978105Z","bundle_sha256":"1b6ebce5cfce0dcdb5b96cd17ff320ba7288f7258e545df433d53315979bcf00"}}