{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:2AVEJ224XGJVZ6B7SRXGKERUT2","short_pith_number":"pith:2AVEJ224","canonical_record":{"source":{"id":"2605.20868","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-20T08:04:40Z","cross_cats_sorted":["cs.AI","cs.SY","eess.SY"],"title_canon_sha256":"0dcf71d246e06d3bfb10f27bea61d882486410cdef4b167215729d76b82c5923","abstract_canon_sha256":"fbf043af13a8b67ce225f5911142002022a21a052e22da9a45ae1174f3ae3fee"},"schema_version":"1.0"},"canonical_sha256":"d02a44eb5cb9935cf83f946e6512349e9446f936618d321b0e8449d212cb3b98","source":{"kind":"arxiv","id":"2605.20868","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.20868","created_at":"2026-05-21T01:05:25Z"},{"alias_kind":"arxiv_version","alias_value":"2605.20868v1","created_at":"2026-05-21T01:05:25Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.20868","created_at":"2026-05-21T01:05:25Z"},{"alias_kind":"pith_short_12","alias_value":"2AVEJ224XGJV","created_at":"2026-05-21T01:05:25Z"},{"alias_kind":"pith_short_16","alias_value":"2AVEJ224XGJVZ6B7","created_at":"2026-05-21T01:05:25Z"},{"alias_kind":"pith_short_8","alias_value":"2AVEJ224","created_at":"2026-05-21T01:05:25Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:2AVEJ224XGJVZ6B7SRXGKERUT2","target":"record","payload":{"canonical_record":{"source":{"id":"2605.20868","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-20T08:04:40Z","cross_cats_sorted":["cs.AI","cs.SY","eess.SY"],"title_canon_sha256":"0dcf71d246e06d3bfb10f27bea61d882486410cdef4b167215729d76b82c5923","abstract_canon_sha256":"fbf043af13a8b67ce225f5911142002022a21a052e22da9a45ae1174f3ae3fee"},"schema_version":"1.0"},"canonical_sha256":"d02a44eb5cb9935cf83f946e6512349e9446f936618d321b0e8449d212cb3b98","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-21T01:05:25.363735Z","signature_b64":"lLuaWoFFhSg48rBIC67jfODee5I0NIu6a3miYqEBlRtgFLG7Es93Jue/w3Fni9IUU18cQwoM8WTcWmTPNCtNBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"d02a44eb5cb9935cf83f946e6512349e9446f936618d321b0e8449d212cb3b98","last_reissued_at":"2026-05-21T01:05:25.363197Z","signature_status":"signed_v1","first_computed_at":"2026-05-21T01:05:25.363197Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2605.20868","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":"2026-05-21T01:05:25Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"rC8aPXUvOJQ5RE+/q7jTqoxa1h6XcyDHovKHGBgYeJrPm3Vkne9KU0be0EkNK6sMmbe37h/qjUrtwKA+iaHmCQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-23T15:12:01.900677Z"},"content_sha256":"a1bbf55ff0bcd0364df028a8706abb911097c333995d485ca129863ce1e84a44","schema_version":"1.0","event_id":"sha256:a1bbf55ff0bcd0364df028a8706abb911097c333995d485ca129863ce1e84a44"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:2AVEJ224XGJVZ6B7SRXGKERUT2","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Runtime-Certified Bounded-Error Quantized Attention","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.SY","eess.SY"],"primary_cat":"cs.LG","authors_text":"Dean Calver","submitted_at":"2026-05-20T08:04:40Z","abstract_excerpt":"KV cache quantization reduces the memory cost of long-context LLM inference, but introduces approximation error that is typically validated only empirically. Existing systems rely on average-case robustness, with no mechanism to detect or recover from failures at runtime. We present a tiered KV cache architecture that enables runtime-certified attention: INT8 keys and INT4 values are stored in GPU memory, while FP16 originals are retained in system RAM for deterministic fallback. A two-term error decomposition yields per-head, per-step bounds on (i) attention distribution distortion from key q"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.20868","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/2605.20868/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":"2026-05-21T01:05:25Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"7c0hYvdbvbigmxdornQacFqhm/8VCey+YqwdSebkeoJNVNDJ0AskcrkbPGNtwc6vRBfgs6D1w57kIaKDv66eDg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-23T15:12:01.901445Z"},"content_sha256":"3c26e2b724af274283e583bbb43ad93bcac3d52e82cbaf0376dc9206b26f6bdd","schema_version":"1.0","event_id":"sha256:3c26e2b724af274283e583bbb43ad93bcac3d52e82cbaf0376dc9206b26f6bdd"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/2AVEJ224XGJVZ6B7SRXGKERUT2/bundle.json","state_url":"https://pith.science/pith/2AVEJ224XGJVZ6B7SRXGKERUT2/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/2AVEJ224XGJVZ6B7SRXGKERUT2/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":"2026-05-23T15:12:01Z","links":{"resolver":"https://pith.science/pith/2AVEJ224XGJVZ6B7SRXGKERUT2","bundle":"https://pith.science/pith/2AVEJ224XGJVZ6B7SRXGKERUT2/bundle.json","state":"https://pith.science/pith/2AVEJ224XGJVZ6B7SRXGKERUT2/state.json","well_known_bundle":"https://pith.science/.well-known/pith/2AVEJ224XGJVZ6B7SRXGKERUT2/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:2AVEJ224XGJVZ6B7SRXGKERUT2","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":"fbf043af13a8b67ce225f5911142002022a21a052e22da9a45ae1174f3ae3fee","cross_cats_sorted":["cs.AI","cs.SY","eess.SY"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-20T08:04:40Z","title_canon_sha256":"0dcf71d246e06d3bfb10f27bea61d882486410cdef4b167215729d76b82c5923"},"schema_version":"1.0","source":{"id":"2605.20868","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.20868","created_at":"2026-05-21T01:05:25Z"},{"alias_kind":"arxiv_version","alias_value":"2605.20868v1","created_at":"2026-05-21T01:05:25Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.20868","created_at":"2026-05-21T01:05:25Z"},{"alias_kind":"pith_short_12","alias_value":"2AVEJ224XGJV","created_at":"2026-05-21T01:05:25Z"},{"alias_kind":"pith_short_16","alias_value":"2AVEJ224XGJVZ6B7","created_at":"2026-05-21T01:05:25Z"},{"alias_kind":"pith_short_8","alias_value":"2AVEJ224","created_at":"2026-05-21T01:05:25Z"}],"graph_snapshots":[{"event_id":"sha256:3c26e2b724af274283e583bbb43ad93bcac3d52e82cbaf0376dc9206b26f6bdd","target":"graph","created_at":"2026-05-21T01:05:25Z","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/2605.20868/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"KV cache quantization reduces the memory cost of long-context LLM inference, but introduces approximation error that is typically validated only empirically. Existing systems rely on average-case robustness, with no mechanism to detect or recover from failures at runtime. We present a tiered KV cache architecture that enables runtime-certified attention: INT8 keys and INT4 values are stored in GPU memory, while FP16 originals are retained in system RAM for deterministic fallback. A two-term error decomposition yields per-head, per-step bounds on (i) attention distribution distortion from key q","authors_text":"Dean Calver","cross_cats":["cs.AI","cs.SY","eess.SY"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-20T08:04:40Z","title":"Runtime-Certified Bounded-Error Quantized Attention"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.20868","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:a1bbf55ff0bcd0364df028a8706abb911097c333995d485ca129863ce1e84a44","target":"record","created_at":"2026-05-21T01:05:25Z","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":"fbf043af13a8b67ce225f5911142002022a21a052e22da9a45ae1174f3ae3fee","cross_cats_sorted":["cs.AI","cs.SY","eess.SY"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-20T08:04:40Z","title_canon_sha256":"0dcf71d246e06d3bfb10f27bea61d882486410cdef4b167215729d76b82c5923"},"schema_version":"1.0","source":{"id":"2605.20868","kind":"arxiv","version":1}},"canonical_sha256":"d02a44eb5cb9935cf83f946e6512349e9446f936618d321b0e8449d212cb3b98","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"d02a44eb5cb9935cf83f946e6512349e9446f936618d321b0e8449d212cb3b98","first_computed_at":"2026-05-21T01:05:25.363197Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-21T01:05:25.363197Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"lLuaWoFFhSg48rBIC67jfODee5I0NIu6a3miYqEBlRtgFLG7Es93Jue/w3Fni9IUU18cQwoM8WTcWmTPNCtNBA==","signature_status":"signed_v1","signed_at":"2026-05-21T01:05:25.363735Z","signed_message":"canonical_sha256_bytes"},"source_id":"2605.20868","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:a1bbf55ff0bcd0364df028a8706abb911097c333995d485ca129863ce1e84a44","sha256:3c26e2b724af274283e583bbb43ad93bcac3d52e82cbaf0376dc9206b26f6bdd"],"state_sha256":"453b463f194c4b6fcc37fa9a4fd42da7fbbe298fcfd303261ec00fb44619ba2b"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"MUvdOWWvrMUaK7yQ32Qq4zgI3O414TmcQ2+/pLAudPI3nxXw6zGS3axiw0egjh9jRNShKkY7ybHJc/cFTUkxBg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-05-23T15:12:01.906092Z","bundle_sha256":"644f36e7e5667269f27487a7d630a96b064205c6fcd5a9e6a1a9ab744e34a03a"}}