{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:A44N37R3E7JJHYWIIDK6E2DAOK","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":"ff33c44ae3235b69dc21aea4bd7667511604cea3837f20abfef65d09166fde4e","cross_cats_sorted":["cs.DC"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-12-04T08:51:23Z","title_canon_sha256":"5245cc2088c6ef08499859b00d9b47e0b305d103c9f3478955b6bc4b14bc1071"},"schema_version":"1.0","source":{"id":"2412.03131","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2412.03131","created_at":"2026-07-05T12:01:56Z"},{"alias_kind":"arxiv_version","alias_value":"2412.03131v3","created_at":"2026-07-05T12:01:56Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2412.03131","created_at":"2026-07-05T12:01:56Z"},{"alias_kind":"pith_short_12","alias_value":"A44N37R3E7JJ","created_at":"2026-07-05T12:01:56Z"},{"alias_kind":"pith_short_16","alias_value":"A44N37R3E7JJHYWI","created_at":"2026-07-05T12:01:56Z"},{"alias_kind":"pith_short_8","alias_value":"A44N37R3","created_at":"2026-07-05T12:01:56Z"}],"graph_snapshots":[{"event_id":"sha256:cb01b835a4d42c32211d25954dfbdd71936d54554abecbaa5bf47df78e59823c","target":"graph","created_at":"2026-07-05T12:01:56Z","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/2412.03131/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Large language models (LLMs) demonstrate remarkable capabilities but face substantial serving costs due to their high memory demands, with the key-value (KV) cache being a primary bottleneck. State-of-the-art KV cache compression techniques, such as quantization and pruning, apply uniform treatment to both keys and values, and discard unimportant tokens entirely, overlooking the fine-grained distinctions in the significance of individual KV cache components. To address such limitations, we introduce \\textit{DiffKV}, a novel framework for efficient KV cache compression that exploits three level","authors_text":"Haibo Chen, John C.S. Lui, Runyuan Zhao, Yanqi Zhang, Yuwei Hu","cross_cats":["cs.DC"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-12-04T08:51:23Z","title":"DiffKV: Differentiated Memory Management for Large Language Models with Parallel KV Compaction"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2412.03131","kind":"arxiv","version":3},"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:dec6c3cb36e79a61ea06459953527414f4e554ee95fb97f0338fe46eb85f26a6","target":"record","created_at":"2026-07-05T12:01:56Z","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":"ff33c44ae3235b69dc21aea4bd7667511604cea3837f20abfef65d09166fde4e","cross_cats_sorted":["cs.DC"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-12-04T08:51:23Z","title_canon_sha256":"5245cc2088c6ef08499859b00d9b47e0b305d103c9f3478955b6bc4b14bc1071"},"schema_version":"1.0","source":{"id":"2412.03131","kind":"arxiv","version":3}},"canonical_sha256":"0738ddfe3b27d293e2c840d5e26860729778bb30852bf8754304f61c15d11e91","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"0738ddfe3b27d293e2c840d5e26860729778bb30852bf8754304f61c15d11e91","first_computed_at":"2026-07-05T12:01:56.164768Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T12:01:56.164768Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"n1MswfL/36XO2xW7RCNH2osvFnL893oukr2Scbxv/JNBghis49zHT5SVy/nfwVAa5qfRR1DgABZjNuC3ooWbDg==","signature_status":"signed_v1","signed_at":"2026-07-05T12:01:56.165276Z","signed_message":"canonical_sha256_bytes"},"source_id":"2412.03131","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:dec6c3cb36e79a61ea06459953527414f4e554ee95fb97f0338fe46eb85f26a6","sha256:cb01b835a4d42c32211d25954dfbdd71936d54554abecbaa5bf47df78e59823c"],"state_sha256":"547c9b18c2c772d59c2e83b3c203170e38628f35d17831341212b07dbc20ae24"}