{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:PZGEYYSIG6XACO42QXQK3TUYUO","short_pith_number":"pith:PZGEYYSI","canonical_record":{"source":{"id":"2503.24000","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-03-31T12:23:31Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"79cd2dfca31918c1b71e6fff49f3cae4d18e675ba88965d2b7aaf59ee32055a3","abstract_canon_sha256":"efee0b6af291b281a0b4b57e766ef8344bd550509fe426fed6b85043ac7be1d6"},"schema_version":"1.0"},"canonical_sha256":"7e4c4c624837ae013b9a85e0adce98a3b9eef0de3c2f1480a46377569ecf2891","source":{"kind":"arxiv","id":"2503.24000","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2503.24000","created_at":"2026-07-05T10:42:02Z"},{"alias_kind":"arxiv_version","alias_value":"2503.24000v1","created_at":"2026-07-05T10:42:02Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2503.24000","created_at":"2026-07-05T10:42:02Z"},{"alias_kind":"pith_short_12","alias_value":"PZGEYYSIG6XA","created_at":"2026-07-05T10:42:02Z"},{"alias_kind":"pith_short_16","alias_value":"PZGEYYSIG6XACO42","created_at":"2026-07-05T10:42:02Z"},{"alias_kind":"pith_short_8","alias_value":"PZGEYYSI","created_at":"2026-07-05T10:42:02Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:PZGEYYSIG6XACO42QXQK3TUYUO","target":"record","payload":{"canonical_record":{"source":{"id":"2503.24000","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-03-31T12:23:31Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"79cd2dfca31918c1b71e6fff49f3cae4d18e675ba88965d2b7aaf59ee32055a3","abstract_canon_sha256":"efee0b6af291b281a0b4b57e766ef8344bd550509fe426fed6b85043ac7be1d6"},"schema_version":"1.0"},"canonical_sha256":"7e4c4c624837ae013b9a85e0adce98a3b9eef0de3c2f1480a46377569ecf2891","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:42:02.988841Z","signature_b64":"v5kN0e+5B9O1/P6iIMhv7g5bxSXLHpkAGzL30n7hy5wB506OuuLjqbUtpSkuslYSkoQd/tiztp0gWAb9JA+8Bg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"7e4c4c624837ae013b9a85e0adce98a3b9eef0de3c2f1480a46377569ecf2891","last_reissued_at":"2026-07-05T10:42:02.988398Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:42:02.988398Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2503.24000","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":"2026-07-05T10:42:02Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Ce7ZvX/OmYBP5KiGRdYyD3gU5G0NL5m52zDNPt1MIEoV48vKU9fPfXy3YPCQwW5T7gsBsMdU+TB23kSmCD0GBQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-21T09:42:45.621582Z"},"content_sha256":"3e1c43408b30592d40fd5b07ecec791af75927c4f9dd97318f3a8d586390bbbd","schema_version":"1.0","event_id":"sha256:3e1c43408b30592d40fd5b07ecec791af75927c4f9dd97318f3a8d586390bbbd"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:PZGEYYSIG6XACO42QXQK3TUYUO","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Rethinking Key-Value Cache Compression Techniques for Large Language Model Serving","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Peng Sun, Tianwei Zhang, Wei Gao, Xinyu Zhou, Yonggang Wen","submitted_at":"2025-03-31T12:23:31Z","abstract_excerpt":"Key-Value cache (\\texttt{KV} \\texttt{cache}) compression has emerged as a promising technique to optimize Large Language Model (LLM) serving. It primarily decreases the memory consumption of \\texttt{KV} \\texttt{cache} to reduce the computation cost. Despite the development of many compression algorithms, their applications in production environments are still not prevalent. In this paper, we revisit mainstream \\texttt{KV} \\texttt{cache} compression solutions from a practical perspective. Our contributions are three-fold. First, we comprehensively review existing algorithmic designs and benchma"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2503.24000","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/2503.24000/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":"2026-07-05T10:42:02Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"V+ygERQNp0m/P6L+ciBH7m5+EObB+jqPxkuWgZQ4Bb8cFqztk48v+U1mvYbVb0L/cAUEsRnEWExHaocNbUaKBw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-21T09:42:45.622086Z"},"content_sha256":"3234e016bb090ec574a977a06b7a1bb88e80c16fea79dd5962ee779e4811e35e","schema_version":"1.0","event_id":"sha256:3234e016bb090ec574a977a06b7a1bb88e80c16fea79dd5962ee779e4811e35e"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/PZGEYYSIG6XACO42QXQK3TUYUO/bundle.json","state_url":"https://pith.science/pith/PZGEYYSIG6XACO42QXQK3TUYUO/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/PZGEYYSIG6XACO42QXQK3TUYUO/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":"2026-08-21T09:42:45Z","links":{"resolver":"https://pith.science/pith/PZGEYYSIG6XACO42QXQK3TUYUO","bundle":"https://pith.science/pith/PZGEYYSIG6XACO42QXQK3TUYUO/bundle.json","state":"https://pith.science/pith/PZGEYYSIG6XACO42QXQK3TUYUO/state.json","well_known_bundle":"https://pith.science/.well-known/pith/PZGEYYSIG6XACO42QXQK3TUYUO/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:PZGEYYSIG6XACO42QXQK3TUYUO","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":"efee0b6af291b281a0b4b57e766ef8344bd550509fe426fed6b85043ac7be1d6","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-03-31T12:23:31Z","title_canon_sha256":"79cd2dfca31918c1b71e6fff49f3cae4d18e675ba88965d2b7aaf59ee32055a3"},"schema_version":"1.0","source":{"id":"2503.24000","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2503.24000","created_at":"2026-07-05T10:42:02Z"},{"alias_kind":"arxiv_version","alias_value":"2503.24000v1","created_at":"2026-07-05T10:42:02Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2503.24000","created_at":"2026-07-05T10:42:02Z"},{"alias_kind":"pith_short_12","alias_value":"PZGEYYSIG6XA","created_at":"2026-07-05T10:42:02Z"},{"alias_kind":"pith_short_16","alias_value":"PZGEYYSIG6XACO42","created_at":"2026-07-05T10:42:02Z"},{"alias_kind":"pith_short_8","alias_value":"PZGEYYSI","created_at":"2026-07-05T10:42:02Z"}],"graph_snapshots":[{"event_id":"sha256:3234e016bb090ec574a977a06b7a1bb88e80c16fea79dd5962ee779e4811e35e","target":"graph","created_at":"2026-07-05T10:42:02Z","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/2503.24000/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Key-Value cache (\\texttt{KV} \\texttt{cache}) compression has emerged as a promising technique to optimize Large Language Model (LLM) serving. It primarily decreases the memory consumption of \\texttt{KV} \\texttt{cache} to reduce the computation cost. Despite the development of many compression algorithms, their applications in production environments are still not prevalent. In this paper, we revisit mainstream \\texttt{KV} \\texttt{cache} compression solutions from a practical perspective. Our contributions are three-fold. First, we comprehensively review existing algorithmic designs and benchma","authors_text":"Peng Sun, Tianwei Zhang, Wei Gao, Xinyu Zhou, Yonggang Wen","cross_cats":["cs.AI"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-03-31T12:23:31Z","title":"Rethinking Key-Value Cache Compression Techniques for Large Language Model Serving"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2503.24000","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":"sha256:3e1c43408b30592d40fd5b07ecec791af75927c4f9dd97318f3a8d586390bbbd","target":"record","created_at":"2026-07-05T10:42:02Z","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":"efee0b6af291b281a0b4b57e766ef8344bd550509fe426fed6b85043ac7be1d6","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-03-31T12:23:31Z","title_canon_sha256":"79cd2dfca31918c1b71e6fff49f3cae4d18e675ba88965d2b7aaf59ee32055a3"},"schema_version":"1.0","source":{"id":"2503.24000","kind":"arxiv","version":1}},"canonical_sha256":"7e4c4c624837ae013b9a85e0adce98a3b9eef0de3c2f1480a46377569ecf2891","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"7e4c4c624837ae013b9a85e0adce98a3b9eef0de3c2f1480a46377569ecf2891","first_computed_at":"2026-07-05T10:42:02.988398Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T10:42:02.988398Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"v5kN0e+5B9O1/P6iIMhv7g5bxSXLHpkAGzL30n7hy5wB506OuuLjqbUtpSkuslYSkoQd/tiztp0gWAb9JA+8Bg==","signature_status":"signed_v1","signed_at":"2026-07-05T10:42:02.988841Z","signed_message":"canonical_sha256_bytes"},"source_id":"2503.24000","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:3e1c43408b30592d40fd5b07ecec791af75927c4f9dd97318f3a8d586390bbbd","sha256:3234e016bb090ec574a977a06b7a1bb88e80c16fea79dd5962ee779e4811e35e"],"state_sha256":"2e3e1ed0f08361357c2518797315cd44057007bd50e9bfbe6b9cffdc7c3e6c5a"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"1+9y4nNrCl6s0XTrDYCmodEnbvOLVP1WC1OnfWhP09x+iIw2SiMuvG+CZmwS2Zr5zwkiETND4d0Whn8ytUB6Dg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-21T09:42:45.627148Z","bundle_sha256":"1fcf9b687abda8c7e8a4799a65bdb2adecd7ded981d0469a4b22ff4836524f65"}}