{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:KDLQGMOTUZSI6QBY35GFQBATCD","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":"2633ba6d614454859b855dd36d921dd278ab6fe19d063aeb6005f4f13bba2e02","cross_cats_sorted":["cs.CL"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-05-21T17:59:29Z","title_canon_sha256":"38e542352699e41ef7bb08fd10f0eb6cf9d6288ed731580ebeac1be21853ee79"},"schema_version":"1.0","source":{"id":"2405.12981","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2405.12981","created_at":"2026-07-05T08:21:29Z"},{"alias_kind":"arxiv_version","alias_value":"2405.12981v1","created_at":"2026-07-05T08:21:29Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2405.12981","created_at":"2026-07-05T08:21:29Z"},{"alias_kind":"pith_short_12","alias_value":"KDLQGMOTUZSI","created_at":"2026-07-05T08:21:29Z"},{"alias_kind":"pith_short_16","alias_value":"KDLQGMOTUZSI6QBY","created_at":"2026-07-05T08:21:29Z"},{"alias_kind":"pith_short_8","alias_value":"KDLQGMOT","created_at":"2026-07-05T08:21:29Z"}],"graph_snapshots":[{"event_id":"sha256:f27bc80d6c26a3256570a354f1fc96667aee4e922361b360110bd3ee8e7613b2","target":"graph","created_at":"2026-07-05T08:21:29Z","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/2405.12981/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Key-value (KV) caching plays an essential role in accelerating decoding for transformer-based autoregressive large language models (LLMs). However, the amount of memory required to store the KV cache can become prohibitive at long sequence lengths and large batch sizes. Since the invention of the transformer, two of the most effective interventions discovered for reducing the size of the KV cache have been Multi-Query Attention (MQA) and its generalization, Grouped-Query Attention (GQA). MQA and GQA both modify the design of the attention block so that multiple query heads can share a single k","authors_text":"Aniruddha Nrusimha, Jonathan Ragan Kelly, Mayank Mishra, Rameswar Panda, William Brandon","cross_cats":["cs.CL"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-05-21T17:59:29Z","title":"Reducing Transformer Key-Value Cache Size with Cross-Layer Attention"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2405.12981","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":"sha256:67d352ecaef91c189a4b72c54059932bfd34aa48decb4f6ada442aa9d0476034","target":"record","created_at":"2026-07-05T08:21:29Z","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":"2633ba6d614454859b855dd36d921dd278ab6fe19d063aeb6005f4f13bba2e02","cross_cats_sorted":["cs.CL"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-05-21T17:59:29Z","title_canon_sha256":"38e542352699e41ef7bb08fd10f0eb6cf9d6288ed731580ebeac1be21853ee79"},"schema_version":"1.0","source":{"id":"2405.12981","kind":"arxiv","version":1}},"canonical_sha256":"50d70331d3a6648f4038df4c58041310f2c3f034603364f049bcaa6d537bfea0","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"50d70331d3a6648f4038df4c58041310f2c3f034603364f049bcaa6d537bfea0","first_computed_at":"2026-07-05T08:21:29.342890Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T08:21:29.342890Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"AYjVYI4KkK+iUeaQzkD4LNCFXa6ALPG39nygDa38g1PfJx8WnPx+LIKAGsyUUk+Olb4Nq3lThpNLnjS061KrAg==","signature_status":"signed_v1","signed_at":"2026-07-05T08:21:29.343363Z","signed_message":"canonical_sha256_bytes"},"source_id":"2405.12981","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:67d352ecaef91c189a4b72c54059932bfd34aa48decb4f6ada442aa9d0476034","sha256:f27bc80d6c26a3256570a354f1fc96667aee4e922361b360110bd3ee8e7613b2"],"state_sha256":"44f63272d8f09dfb37feebb1d6d0a13807de61886815dee5562cf0f1524463c0"}