{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:ABSH3VJSHK2ZSCINU2ZODWEKLT","short_pith_number":"pith:ABSH3VJS","canonical_record":{"source":{"id":"2501.19392","kind":"arxiv","version":4},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-01-31T18:47:42Z","cross_cats_sorted":[],"title_canon_sha256":"965422ea026d169eb6768c0b621052f73beebb2a2a1c9c5a6ca830adf3429141","abstract_canon_sha256":"18075049eee8cc56855753f5c5fbc763eccb54eb0327eb86099ae14ee9876003"},"schema_version":"1.0"},"canonical_sha256":"00647dd5323ab599090da6b2e1d88a5cec04cb3bf8f8d81a682c5e1a674fe82a","source":{"kind":"arxiv","id":"2501.19392","version":4},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2501.19392","created_at":"2026-07-05T10:21:52Z"},{"alias_kind":"arxiv_version","alias_value":"2501.19392v4","created_at":"2026-07-05T10:21:52Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2501.19392","created_at":"2026-07-05T10:21:52Z"},{"alias_kind":"pith_short_12","alias_value":"ABSH3VJSHK2Z","created_at":"2026-07-05T10:21:52Z"},{"alias_kind":"pith_short_16","alias_value":"ABSH3VJSHK2ZSCIN","created_at":"2026-07-05T10:21:52Z"},{"alias_kind":"pith_short_8","alias_value":"ABSH3VJS","created_at":"2026-07-05T10:21:52Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:ABSH3VJSHK2ZSCINU2ZODWEKLT","target":"record","payload":{"canonical_record":{"source":{"id":"2501.19392","kind":"arxiv","version":4},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-01-31T18:47:42Z","cross_cats_sorted":[],"title_canon_sha256":"965422ea026d169eb6768c0b621052f73beebb2a2a1c9c5a6ca830adf3429141","abstract_canon_sha256":"18075049eee8cc56855753f5c5fbc763eccb54eb0327eb86099ae14ee9876003"},"schema_version":"1.0"},"canonical_sha256":"00647dd5323ab599090da6b2e1d88a5cec04cb3bf8f8d81a682c5e1a674fe82a","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:21:52.269006Z","signature_b64":"/j+bjVfKcJ7BlxySHaT+JN07Y12XjDJZUCrLYUdqsS1caQboY+3/9fBWukDoinPfA3D/n/FKGrR4n/VfSrNjDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"00647dd5323ab599090da6b2e1d88a5cec04cb3bf8f8d81a682c5e1a674fe82a","last_reissued_at":"2026-07-05T10:21:52.268493Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:21:52.268493Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2501.19392","source_version":4,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T10:21:52Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"duliecQKAwyUkZPU6uNDxx40Ms1zVm/R9zSPuFPxkBu62OgfJgXNYuyrffSfLOGGpO0N+ZeMBuv4j+UoUMW+Bg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-18T05:05:39.841282Z"},"content_sha256":"c711f5cba2b56be000960bbb362833725213db96e87ce73187467880e664b83c","schema_version":"1.0","event_id":"sha256:c711f5cba2b56be000960bbb362833725213db96e87ce73187467880e664b83c"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:ABSH3VJSHK2ZSCINU2ZODWEKLT","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Cache Me If You Must: Adaptive Key-Value Quantization for Large Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Alina Shutova, Dan Alistarh, Denis Kuznedelev, Denis Mazur, Ivan Ermakov, Nikita Surkov, Vage Egiazarian, Vladimir Malinovskii","submitted_at":"2025-01-31T18:47:42Z","abstract_excerpt":"Efficient real-world deployments of large language models (LLMs) rely on Key-Value (KV) caching for processing and generating long outputs, reducing the need for repetitive computation. For large contexts, Key-Value caches can take up tens of gigabytes of device memory, as they store vector representations for each token and layer. Recent work has shown that the cached vectors can be compressed through quantization, pruning or merging, but these techniques often compromise quality towards higher compression rates. In this work, we aim to improve Key & Value compression by exploiting two observ"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2501.19392","kind":"arxiv","version":4},"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/2501.19392/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":"2026-07-05T10:21:52Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"f76YwXy69MAIDSPZfbUX575r9bDIGoX8NBWPH/DqeS+v9UtnenLUGcM4GvBA9ec82L2qjHtA5KHt5HsG/1+3AQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-18T05:05:39.841656Z"},"content_sha256":"c1be12e4db5b6dfbc73c3ade6edb7e4bba7f62e03af7d3b4163d7ff70bd7c5bf","schema_version":"1.0","event_id":"sha256:c1be12e4db5b6dfbc73c3ade6edb7e4bba7f62e03af7d3b4163d7ff70bd7c5bf"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/ABSH3VJSHK2ZSCINU2ZODWEKLT/bundle.json","state_url":"https://pith.science/pith/ABSH3VJSHK2ZSCINU2ZODWEKLT/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/ABSH3VJSHK2ZSCINU2ZODWEKLT/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":"2026-07-18T05:05:39Z","links":{"resolver":"https://pith.science/pith/ABSH3VJSHK2ZSCINU2ZODWEKLT","bundle":"https://pith.science/pith/ABSH3VJSHK2ZSCINU2ZODWEKLT/bundle.json","state":"https://pith.science/pith/ABSH3VJSHK2ZSCINU2ZODWEKLT/state.json","well_known_bundle":"https://pith.science/.well-known/pith/ABSH3VJSHK2ZSCINU2ZODWEKLT/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:ABSH3VJSHK2ZSCINU2ZODWEKLT","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":"18075049eee8cc56855753f5c5fbc763eccb54eb0327eb86099ae14ee9876003","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-01-31T18:47:42Z","title_canon_sha256":"965422ea026d169eb6768c0b621052f73beebb2a2a1c9c5a6ca830adf3429141"},"schema_version":"1.0","source":{"id":"2501.19392","kind":"arxiv","version":4}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2501.19392","created_at":"2026-07-05T10:21:52Z"},{"alias_kind":"arxiv_version","alias_value":"2501.19392v4","created_at":"2026-07-05T10:21:52Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2501.19392","created_at":"2026-07-05T10:21:52Z"},{"alias_kind":"pith_short_12","alias_value":"ABSH3VJSHK2Z","created_at":"2026-07-05T10:21:52Z"},{"alias_kind":"pith_short_16","alias_value":"ABSH3VJSHK2ZSCIN","created_at":"2026-07-05T10:21:52Z"},{"alias_kind":"pith_short_8","alias_value":"ABSH3VJS","created_at":"2026-07-05T10:21:52Z"}],"graph_snapshots":[{"event_id":"sha256:c1be12e4db5b6dfbc73c3ade6edb7e4bba7f62e03af7d3b4163d7ff70bd7c5bf","target":"graph","created_at":"2026-07-05T10:21:52Z","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/2501.19392/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Efficient real-world deployments of large language models (LLMs) rely on Key-Value (KV) caching for processing and generating long outputs, reducing the need for repetitive computation. For large contexts, Key-Value caches can take up tens of gigabytes of device memory, as they store vector representations for each token and layer. Recent work has shown that the cached vectors can be compressed through quantization, pruning or merging, but these techniques often compromise quality towards higher compression rates. In this work, we aim to improve Key & Value compression by exploiting two observ","authors_text":"Alina Shutova, Dan Alistarh, Denis Kuznedelev, Denis Mazur, Ivan Ermakov, Nikita Surkov, Vage Egiazarian, Vladimir Malinovskii","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-01-31T18:47:42Z","title":"Cache Me If You Must: Adaptive Key-Value Quantization for Large Language Models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2501.19392","kind":"arxiv","version":4},"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:c711f5cba2b56be000960bbb362833725213db96e87ce73187467880e664b83c","target":"record","created_at":"2026-07-05T10:21:52Z","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":"18075049eee8cc56855753f5c5fbc763eccb54eb0327eb86099ae14ee9876003","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-01-31T18:47:42Z","title_canon_sha256":"965422ea026d169eb6768c0b621052f73beebb2a2a1c9c5a6ca830adf3429141"},"schema_version":"1.0","source":{"id":"2501.19392","kind":"arxiv","version":4}},"canonical_sha256":"00647dd5323ab599090da6b2e1d88a5cec04cb3bf8f8d81a682c5e1a674fe82a","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"00647dd5323ab599090da6b2e1d88a5cec04cb3bf8f8d81a682c5e1a674fe82a","first_computed_at":"2026-07-05T10:21:52.268493Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T10:21:52.268493Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"/j+bjVfKcJ7BlxySHaT+JN07Y12XjDJZUCrLYUdqsS1caQboY+3/9fBWukDoinPfA3D/n/FKGrR4n/VfSrNjDQ==","signature_status":"signed_v1","signed_at":"2026-07-05T10:21:52.269006Z","signed_message":"canonical_sha256_bytes"},"source_id":"2501.19392","source_kind":"arxiv","source_version":4}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:c711f5cba2b56be000960bbb362833725213db96e87ce73187467880e664b83c","sha256:c1be12e4db5b6dfbc73c3ade6edb7e4bba7f62e03af7d3b4163d7ff70bd7c5bf"],"state_sha256":"d4e10e0636addb678946e7833d3d435cbf3d21a95acafea8ac191b7ed6757eb6"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"WCktN1GbHBMNawDHh6WnDLpZAODqXZ/ZCCaehyeSibnVuWB6UV14qnJbXcXwJtAXrzqavdiKVasiV1WgZylDBA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-18T05:05:39.843574Z","bundle_sha256":"fd271bc23b99df4278067996df7fc64be42ed48a2665908a63fd98ee28977dcb"}}