{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:ZKGBG2ZCLP4VQ534CSTM3KNPS5","short_pith_number":"pith:ZKGBG2ZC","canonical_record":{"source":{"id":"2401.18079","kind":"arxiv","version":6},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-01-31T18:58:14Z","cross_cats_sorted":[],"title_canon_sha256":"8093b554e94f847fdf86355a6fc1785b19344f5fed33a8a634eefd2b4794cd4e","abstract_canon_sha256":"777a4a4bc787e4fb21f6caaab70dbbdf32806629d57e4442de0190a263ce18b7"},"schema_version":"1.0"},"canonical_sha256":"ca8c136b225bf958777c14a6cda9af977045e9283a19d387a00110255b24f809","source":{"kind":"arxiv","id":"2401.18079","version":6},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2401.18079","created_at":"2026-07-05T11:11:30Z"},{"alias_kind":"arxiv_version","alias_value":"2401.18079v6","created_at":"2026-07-05T11:11:30Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2401.18079","created_at":"2026-07-05T11:11:30Z"},{"alias_kind":"pith_short_12","alias_value":"ZKGBG2ZCLP4V","created_at":"2026-07-05T11:11:30Z"},{"alias_kind":"pith_short_16","alias_value":"ZKGBG2ZCLP4VQ534","created_at":"2026-07-05T11:11:30Z"},{"alias_kind":"pith_short_8","alias_value":"ZKGBG2ZC","created_at":"2026-07-05T11:11:30Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:ZKGBG2ZCLP4VQ534CSTM3KNPS5","target":"record","payload":{"canonical_record":{"source":{"id":"2401.18079","kind":"arxiv","version":6},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-01-31T18:58:14Z","cross_cats_sorted":[],"title_canon_sha256":"8093b554e94f847fdf86355a6fc1785b19344f5fed33a8a634eefd2b4794cd4e","abstract_canon_sha256":"777a4a4bc787e4fb21f6caaab70dbbdf32806629d57e4442de0190a263ce18b7"},"schema_version":"1.0"},"canonical_sha256":"ca8c136b225bf958777c14a6cda9af977045e9283a19d387a00110255b24f809","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:11:30.840105Z","signature_b64":"HiexR79w3UONsmauWodfAZaVpwOhwjG/0+JodsUV9ZacgdtbwnqlsEzdEmBwvXDHgjibg8kMSlgkq19W/ynIBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"ca8c136b225bf958777c14a6cda9af977045e9283a19d387a00110255b24f809","last_reissued_at":"2026-07-05T11:11:30.838685Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:11:30.838685Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2401.18079","source_version":6,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T11:11:30Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"NP3uyTz5Ea++mD/BgTl+HFzo5kZTloGSBN/boSSOzHE9n7HpzV4zksd2OVM49MPGqN/+VdliJSRbhU50/K8KBw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-03T21:19:43.122896Z"},"content_sha256":"e511fff44ddc89211720e8ef97db8392cb5149fd20d8935645d5043de4e984f8","schema_version":"1.0","event_id":"sha256:e511fff44ddc89211720e8ef97db8392cb5149fd20d8935645d5043de4e984f8"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:ZKGBG2ZCLP4VQ534CSTM3KNPS5","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"KVQuant: Towards 10 Million Context Length LLM Inference with KV Cache Quantization","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Amir Gholami, Coleman Hooper, Hiva Mohammadzadeh, Kurt Keutzer, Michael W. Mahoney, Sehoon Kim, Yakun Sophia Shao","submitted_at":"2024-01-31T18:58:14Z","abstract_excerpt":"LLMs are seeing growing use for applications which require large context windows, and with these large context windows KV cache activations surface as the dominant contributor to memory consumption during inference. Quantization is a promising approach for compressing KV cache activations; however, existing solutions fail to represent activations accurately in sub-4-bit precision. Our work, KVQuant, facilitates low precision KV cache quantization by incorporating several novel methods: (i) Per-Channel Key Quantization, where we adjust the dimension along which we quantize the Key activations t"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2401.18079","kind":"arxiv","version":6},"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/2401.18079/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":"2026-07-05T11:11:30Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"d2XtGQSzOzIa325xIPQy19BlK09bCSOk9/3IwVmpAGN4E5x7mY5lOyJDec+Nq95D/HTogIqEGo433DM/WfhqAw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-03T21:19:43.123545Z"},"content_sha256":"0633becd45abd381a90e277a157a903017bb457287abd91ab29b6d121000f140","schema_version":"1.0","event_id":"sha256:0633becd45abd381a90e277a157a903017bb457287abd91ab29b6d121000f140"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/ZKGBG2ZCLP4VQ534CSTM3KNPS5/bundle.json","state_url":"https://pith.science/pith/ZKGBG2ZCLP4VQ534CSTM3KNPS5/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/ZKGBG2ZCLP4VQ534CSTM3KNPS5/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":"2026-08-03T21:19:43Z","links":{"resolver":"https://pith.science/pith/ZKGBG2ZCLP4VQ534CSTM3KNPS5","bundle":"https://pith.science/pith/ZKGBG2ZCLP4VQ534CSTM3KNPS5/bundle.json","state":"https://pith.science/pith/ZKGBG2ZCLP4VQ534CSTM3KNPS5/state.json","well_known_bundle":"https://pith.science/.well-known/pith/ZKGBG2ZCLP4VQ534CSTM3KNPS5/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:ZKGBG2ZCLP4VQ534CSTM3KNPS5","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":"777a4a4bc787e4fb21f6caaab70dbbdf32806629d57e4442de0190a263ce18b7","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-01-31T18:58:14Z","title_canon_sha256":"8093b554e94f847fdf86355a6fc1785b19344f5fed33a8a634eefd2b4794cd4e"},"schema_version":"1.0","source":{"id":"2401.18079","kind":"arxiv","version":6}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2401.18079","created_at":"2026-07-05T11:11:30Z"},{"alias_kind":"arxiv_version","alias_value":"2401.18079v6","created_at":"2026-07-05T11:11:30Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2401.18079","created_at":"2026-07-05T11:11:30Z"},{"alias_kind":"pith_short_12","alias_value":"ZKGBG2ZCLP4V","created_at":"2026-07-05T11:11:30Z"},{"alias_kind":"pith_short_16","alias_value":"ZKGBG2ZCLP4VQ534","created_at":"2026-07-05T11:11:30Z"},{"alias_kind":"pith_short_8","alias_value":"ZKGBG2ZC","created_at":"2026-07-05T11:11:30Z"}],"graph_snapshots":[{"event_id":"sha256:0633becd45abd381a90e277a157a903017bb457287abd91ab29b6d121000f140","target":"graph","created_at":"2026-07-05T11:11:30Z","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/2401.18079/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"LLMs are seeing growing use for applications which require large context windows, and with these large context windows KV cache activations surface as the dominant contributor to memory consumption during inference. Quantization is a promising approach for compressing KV cache activations; however, existing solutions fail to represent activations accurately in sub-4-bit precision. Our work, KVQuant, facilitates low precision KV cache quantization by incorporating several novel methods: (i) Per-Channel Key Quantization, where we adjust the dimension along which we quantize the Key activations t","authors_text":"Amir Gholami, Coleman Hooper, Hiva Mohammadzadeh, Kurt Keutzer, Michael W. Mahoney, Sehoon Kim, Yakun Sophia Shao","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-01-31T18:58:14Z","title":"KVQuant: Towards 10 Million Context Length LLM Inference with KV Cache Quantization"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2401.18079","kind":"arxiv","version":6},"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:e511fff44ddc89211720e8ef97db8392cb5149fd20d8935645d5043de4e984f8","target":"record","created_at":"2026-07-05T11:11:30Z","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":"777a4a4bc787e4fb21f6caaab70dbbdf32806629d57e4442de0190a263ce18b7","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-01-31T18:58:14Z","title_canon_sha256":"8093b554e94f847fdf86355a6fc1785b19344f5fed33a8a634eefd2b4794cd4e"},"schema_version":"1.0","source":{"id":"2401.18079","kind":"arxiv","version":6}},"canonical_sha256":"ca8c136b225bf958777c14a6cda9af977045e9283a19d387a00110255b24f809","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"ca8c136b225bf958777c14a6cda9af977045e9283a19d387a00110255b24f809","first_computed_at":"2026-07-05T11:11:30.838685Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:11:30.838685Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"HiexR79w3UONsmauWodfAZaVpwOhwjG/0+JodsUV9ZacgdtbwnqlsEzdEmBwvXDHgjibg8kMSlgkq19W/ynIBQ==","signature_status":"signed_v1","signed_at":"2026-07-05T11:11:30.840105Z","signed_message":"canonical_sha256_bytes"},"source_id":"2401.18079","source_kind":"arxiv","source_version":6}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:e511fff44ddc89211720e8ef97db8392cb5149fd20d8935645d5043de4e984f8","sha256:0633becd45abd381a90e277a157a903017bb457287abd91ab29b6d121000f140"],"state_sha256":"17c4e34017b953d9eaca6cbfcb4679ef4bf9ca02b383e91af6ee3ec7cda6d249"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"rN9P7cyxsRWeWBSwp3SuY3Fm0aoCKjorQ0AeMCcOwK9Mjv/x8vm5axs+5b5C4JK9a0iy/3MZmtp1sYdnBR8/Ag==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-03T21:19:43.127789Z","bundle_sha256":"49080da9c2ea4810d977985454d89a7b89ffe9599e4a3f9373ea63a08419e74e"}}