{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:HP3FF5LWAKHGYTBD4LDPDZPYAC","short_pith_number":"pith:HP3FF5LW","canonical_record":{"source":{"id":"2509.04377","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-09-04T16:40:01Z","cross_cats_sorted":[],"title_canon_sha256":"ae9555f3d4ff840398432ad4313e595ed4de5e4f8e224c304183e99041f255c4","abstract_canon_sha256":"d5468091c4972ad7893665fe1d4e87b72f79c3b78f5766fc85485afdfe281a2f"},"schema_version":"1.0"},"canonical_sha256":"3bf652f576028e6c4c23e2c6f1e5f80093548e6b67740ae29a5229aa999f777d","source":{"kind":"arxiv","id":"2509.04377","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2509.04377","created_at":"2026-07-05T12:04:58Z"},{"alias_kind":"arxiv_version","alias_value":"2509.04377v1","created_at":"2026-07-05T12:04:58Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2509.04377","created_at":"2026-07-05T12:04:58Z"},{"alias_kind":"pith_short_12","alias_value":"HP3FF5LWAKHG","created_at":"2026-07-05T12:04:58Z"},{"alias_kind":"pith_short_16","alias_value":"HP3FF5LWAKHGYTBD","created_at":"2026-07-05T12:04:58Z"},{"alias_kind":"pith_short_8","alias_value":"HP3FF5LW","created_at":"2026-07-05T12:04:58Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:HP3FF5LWAKHGYTBD4LDPDZPYAC","target":"record","payload":{"canonical_record":{"source":{"id":"2509.04377","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-09-04T16:40:01Z","cross_cats_sorted":[],"title_canon_sha256":"ae9555f3d4ff840398432ad4313e595ed4de5e4f8e224c304183e99041f255c4","abstract_canon_sha256":"d5468091c4972ad7893665fe1d4e87b72f79c3b78f5766fc85485afdfe281a2f"},"schema_version":"1.0"},"canonical_sha256":"3bf652f576028e6c4c23e2c6f1e5f80093548e6b67740ae29a5229aa999f777d","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T12:04:58.550710Z","signature_b64":"hvwwIImtwa1SjXVBIokhLyWT/EYdOyQW9mp3eT4FUXeeVmW1ci/PBk82YlS5p4nahbMNA5JBzDC2UtWhGymrCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"3bf652f576028e6c4c23e2c6f1e5f80093548e6b67740ae29a5229aa999f777d","last_reissued_at":"2026-07-05T12:04:58.550156Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T12:04:58.550156Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2509.04377","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":"2026-07-05T12:04:58Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"3IK2U7KFfxXD8zJ66Ca2MfELf8K3ZBOZikgQmbg4luEI73cQTRhKGp963xVf73TlXyXBu8wvR9+qdsduzqqiBg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-05T17:12:20.736329Z"},"content_sha256":"55e5ecb4876366149b40c508f68470577c968e41c568578606498f6b2a071c43","schema_version":"1.0","event_id":"sha256:55e5ecb4876366149b40c508f68470577c968e41c568578606498f6b2a071c43"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:HP3FF5LWAKHGYTBD4LDPDZPYAC","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"PagedEviction: Structured Block-wise KV Cache Pruning for Efficient Large Language Model Inference","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Anthony Kougkas, Bogdan Nicolae, Jie Ye, Krishna Teja Chitty-Venkata, Murali Emani, Venkatram Vishwanath, Xian-He Sun","submitted_at":"2025-09-04T16:40:01Z","abstract_excerpt":"KV caching significantly improves the efficiency of Large Language Model (LLM) inference by storing attention states from previously processed tokens, enabling faster generation of subsequent tokens. However, as sequence length increases, the KV cache quickly becomes a major memory bottleneck. To address this, we propose PagedEviction, a novel fine-grained, structured KV cache pruning strategy that enhances the memory efficiency of vLLM's PagedAttention. Unlike existing approaches that rely on attention-based token importance or evict tokens across different vLLM pages, PagedEviction introduce"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2509.04377","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/2509.04377/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":"2026-07-05T12:04:58Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"vDGhT+E8h+iwp2/llnjrzX5/8EDoM6FPjW4MEWC5E/KSxcBFdQIUZ0BE815jK9/TpZ7X8cDzzaRPstQaPQvwDQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-05T17:12:20.736713Z"},"content_sha256":"b140cfcd24bfa6be5dcc25176b5b335ff7b73fc23a13bb8efffec1e56effce53","schema_version":"1.0","event_id":"sha256:b140cfcd24bfa6be5dcc25176b5b335ff7b73fc23a13bb8efffec1e56effce53"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/HP3FF5LWAKHGYTBD4LDPDZPYAC/bundle.json","state_url":"https://pith.science/pith/HP3FF5LWAKHGYTBD4LDPDZPYAC/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/HP3FF5LWAKHGYTBD4LDPDZPYAC/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":"2026-08-05T17:12:20Z","links":{"resolver":"https://pith.science/pith/HP3FF5LWAKHGYTBD4LDPDZPYAC","bundle":"https://pith.science/pith/HP3FF5LWAKHGYTBD4LDPDZPYAC/bundle.json","state":"https://pith.science/pith/HP3FF5LWAKHGYTBD4LDPDZPYAC/state.json","well_known_bundle":"https://pith.science/.well-known/pith/HP3FF5LWAKHGYTBD4LDPDZPYAC/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:HP3FF5LWAKHGYTBD4LDPDZPYAC","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":"d5468091c4972ad7893665fe1d4e87b72f79c3b78f5766fc85485afdfe281a2f","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-09-04T16:40:01Z","title_canon_sha256":"ae9555f3d4ff840398432ad4313e595ed4de5e4f8e224c304183e99041f255c4"},"schema_version":"1.0","source":{"id":"2509.04377","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2509.04377","created_at":"2026-07-05T12:04:58Z"},{"alias_kind":"arxiv_version","alias_value":"2509.04377v1","created_at":"2026-07-05T12:04:58Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2509.04377","created_at":"2026-07-05T12:04:58Z"},{"alias_kind":"pith_short_12","alias_value":"HP3FF5LWAKHG","created_at":"2026-07-05T12:04:58Z"},{"alias_kind":"pith_short_16","alias_value":"HP3FF5LWAKHGYTBD","created_at":"2026-07-05T12:04:58Z"},{"alias_kind":"pith_short_8","alias_value":"HP3FF5LW","created_at":"2026-07-05T12:04:58Z"}],"graph_snapshots":[{"event_id":"sha256:b140cfcd24bfa6be5dcc25176b5b335ff7b73fc23a13bb8efffec1e56effce53","target":"graph","created_at":"2026-07-05T12:04:58Z","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/2509.04377/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"KV caching significantly improves the efficiency of Large Language Model (LLM) inference by storing attention states from previously processed tokens, enabling faster generation of subsequent tokens. However, as sequence length increases, the KV cache quickly becomes a major memory bottleneck. To address this, we propose PagedEviction, a novel fine-grained, structured KV cache pruning strategy that enhances the memory efficiency of vLLM's PagedAttention. Unlike existing approaches that rely on attention-based token importance or evict tokens across different vLLM pages, PagedEviction introduce","authors_text":"Anthony Kougkas, Bogdan Nicolae, Jie Ye, Krishna Teja Chitty-Venkata, Murali Emani, Venkatram Vishwanath, Xian-He Sun","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-09-04T16:40:01Z","title":"PagedEviction: Structured Block-wise KV Cache Pruning for Efficient Large Language Model Inference"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2509.04377","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":"sha256:55e5ecb4876366149b40c508f68470577c968e41c568578606498f6b2a071c43","target":"record","created_at":"2026-07-05T12:04:58Z","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":"d5468091c4972ad7893665fe1d4e87b72f79c3b78f5766fc85485afdfe281a2f","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-09-04T16:40:01Z","title_canon_sha256":"ae9555f3d4ff840398432ad4313e595ed4de5e4f8e224c304183e99041f255c4"},"schema_version":"1.0","source":{"id":"2509.04377","kind":"arxiv","version":1}},"canonical_sha256":"3bf652f576028e6c4c23e2c6f1e5f80093548e6b67740ae29a5229aa999f777d","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"3bf652f576028e6c4c23e2c6f1e5f80093548e6b67740ae29a5229aa999f777d","first_computed_at":"2026-07-05T12:04:58.550156Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T12:04:58.550156Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"hvwwIImtwa1SjXVBIokhLyWT/EYdOyQW9mp3eT4FUXeeVmW1ci/PBk82YlS5p4nahbMNA5JBzDC2UtWhGymrCw==","signature_status":"signed_v1","signed_at":"2026-07-05T12:04:58.550710Z","signed_message":"canonical_sha256_bytes"},"source_id":"2509.04377","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:55e5ecb4876366149b40c508f68470577c968e41c568578606498f6b2a071c43","sha256:b140cfcd24bfa6be5dcc25176b5b335ff7b73fc23a13bb8efffec1e56effce53"],"state_sha256":"4fb0ab61aeb9bfff144b69be92fda578db6146a1e19070f6c78b9b1e20823e85"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"B7/S5G36bJn46NjHw6MRBBgALm++rsUdN+WHE4/owSnN/aXLvV1POUcDVjy0+eyc8O+y5xA7YaLe8Rao9A6bCw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-05T17:12:20.744667Z","bundle_sha256":"acc5aea0f5b0702f62aa3e4adc6590af6b8123b31d2037b7fae2d96071c5de9a"}}