{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:QQQUQSKZWJ7IXWYBW6HBGALYHD","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":"0ea8acb4b3c4e8f8fab949af65e23c586b3fefbcc8043c80f081d31b16b2be22","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.LG","submitted_at":"2024-06-13T16:33:44Z","title_canon_sha256":"aaea7180f1940b3bc68d8b310d6345b181cfdddd2a535492fc111b49744e300b"},"schema_version":"1.0","source":{"id":"2406.09297","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2406.09297","created_at":"2026-07-05T09:20:44Z"},{"alias_kind":"arxiv_version","alias_value":"2406.09297v3","created_at":"2026-07-05T09:20:44Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2406.09297","created_at":"2026-07-05T09:20:44Z"},{"alias_kind":"pith_short_12","alias_value":"QQQUQSKZWJ7I","created_at":"2026-07-05T09:20:44Z"},{"alias_kind":"pith_short_16","alias_value":"QQQUQSKZWJ7IXWYB","created_at":"2026-07-05T09:20:44Z"},{"alias_kind":"pith_short_8","alias_value":"QQQUQSKZ","created_at":"2026-07-05T09:20:44Z"}],"graph_snapshots":[{"event_id":"sha256:d7831d5bc5a31ea8f82dd76a136d7c86227ae2101ad5bb4889d0f7aa7792c13e","target":"graph","created_at":"2026-07-05T09:20:44Z","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/2406.09297/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Auto-regressive inference of transformers benefit greatly from Key-Value (KV) caching, but can lead to major memory bottlenecks as model size, batch size, and sequence length grow at scale. We introduce Multi-Layer Key-Value (MLKV) sharing, a novel approach extending KV sharing across transformer layers to reduce memory usage beyond what was possible with Multi-Query Attention (MQA) and Grouped-Query Attention (GQA). Evaluations on various NLP benchmarks and inference metrics using uptrained Pythia-160M variants demonstrate that MLKV significantly reduces memory usage with minimal performance ","authors_text":"Alham Fikri Aji, Ayu Purwarianti, Muhammad Farid Adilazuarda, Zayd Muhammad Kawakibi Zuhri","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.LG","submitted_at":"2024-06-13T16:33:44Z","title":"MLKV: Multi-Layer Key-Value Heads for Memory Efficient Transformer Decoding"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2406.09297","kind":"arxiv","version":3},"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":"sha256:40e0659e36f4000c3c41c2d6172e1bf1416de759707a3d253558fdf17d1a05ac","target":"record","created_at":"2026-07-05T09:20:44Z","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":"0ea8acb4b3c4e8f8fab949af65e23c586b3fefbcc8043c80f081d31b16b2be22","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.LG","submitted_at":"2024-06-13T16:33:44Z","title_canon_sha256":"aaea7180f1940b3bc68d8b310d6345b181cfdddd2a535492fc111b49744e300b"},"schema_version":"1.0","source":{"id":"2406.09297","kind":"arxiv","version":3}},"canonical_sha256":"8421484959b27e8bdb01b78e13017838cf2aebdd9369953810557d4a6e9e3140","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"8421484959b27e8bdb01b78e13017838cf2aebdd9369953810557d4a6e9e3140","first_computed_at":"2026-07-05T09:20:44.238764Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T09:20:44.238764Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"E0ufWdiF6yoZcsN63HcLdEzzthVPnmau5tJMNzOz+ZSRaQfyy1Fie2cfZxIvGoHgNMs27lmAmLPVEiaqbXF5BQ==","signature_status":"signed_v1","signed_at":"2026-07-05T09:20:44.239213Z","signed_message":"canonical_sha256_bytes"},"source_id":"2406.09297","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:40e0659e36f4000c3c41c2d6172e1bf1416de759707a3d253558fdf17d1a05ac","sha256:d7831d5bc5a31ea8f82dd76a136d7c86227ae2101ad5bb4889d0f7aa7792c13e"],"state_sha256":"af08fe9a81697897af9e6a9c6f4139d15ff3d9a0973b1407dfb307be20f33c95"}