{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:UPKZGQLXP4YLYCZXSWYVVY2A74","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":"677f23bf8a97d3bc361e2e6a428354ac02b43fe7580f96543bcf3e3c651f3dd4","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-03-14T17:59:26Z","title_canon_sha256":"e56eb19347be8e219fb34002b760bd962e3ad8186676f4b706c09523d769d08f"},"schema_version":"1.0","source":{"id":"2403.09636","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2403.09636","created_at":"2026-07-05T08:47:16Z"},{"alias_kind":"arxiv_version","alias_value":"2403.09636v2","created_at":"2026-07-05T08:47:16Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2403.09636","created_at":"2026-07-05T08:47:16Z"},{"alias_kind":"pith_short_12","alias_value":"UPKZGQLXP4YL","created_at":"2026-07-05T08:47:16Z"},{"alias_kind":"pith_short_16","alias_value":"UPKZGQLXP4YLYCZX","created_at":"2026-07-05T08:47:16Z"},{"alias_kind":"pith_short_8","alias_value":"UPKZGQLX","created_at":"2026-07-05T08:47:16Z"}],"graph_snapshots":[{"event_id":"sha256:24f52ba9435065dc6b2039b3d209b1e31853f2b6ad3d16b30f2cfe71de61f2fe","target":"graph","created_at":"2026-07-05T08:47:16Z","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/2403.09636/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Transformers have emerged as the backbone of large language models (LLMs). However, generation remains inefficient due to the need to store in memory a cache of key-value representations for past tokens, whose size scales linearly with the input sequence length and batch size. As a solution, we propose Dynamic Memory Compression (DMC), a method for online key-value cache compression at inference time. Most importantly, the model learns to apply different compression ratios in different heads and layers. We retrofit pre-trained LLMs such as Llama 2 (7B, 13B and 70B) into DMC Transformers, achie","authors_text":"Adrian {\\L}a\\'ncucki, David Tarjan, Edoardo M. Ponti, Marcin Chochowski, Piotr Nawrot","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-03-14T17:59:26Z","title":"Dynamic Memory Compression: Retrofitting LLMs for Accelerated Inference"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2403.09636","kind":"arxiv","version":2},"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:d978f5ff901831796cf17864493458a3e68361f363f6fddac75c60c052190201","target":"record","created_at":"2026-07-05T08:47:16Z","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":"677f23bf8a97d3bc361e2e6a428354ac02b43fe7580f96543bcf3e3c651f3dd4","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-03-14T17:59:26Z","title_canon_sha256":"e56eb19347be8e219fb34002b760bd962e3ad8186676f4b706c09523d769d08f"},"schema_version":"1.0","source":{"id":"2403.09636","kind":"arxiv","version":2}},"canonical_sha256":"a3d59341777f30bc0b3795b15ae340ff1a3c8f866c46038c60bbabc2efc1e926","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"a3d59341777f30bc0b3795b15ae340ff1a3c8f866c46038c60bbabc2efc1e926","first_computed_at":"2026-07-05T08:47:16.653023Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T08:47:16.653023Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"Tb4tM+VchLtkCKCvtkWK/K3KQiIrqeTRE+U2MeOyx7TG4IdvQMk7IuNHj6pgAtBGO1KquYOj2nST7ANJ3pfiAw==","signature_status":"signed_v1","signed_at":"2026-07-05T08:47:16.653518Z","signed_message":"canonical_sha256_bytes"},"source_id":"2403.09636","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:d978f5ff901831796cf17864493458a3e68361f363f6fddac75c60c052190201","sha256:24f52ba9435065dc6b2039b3d209b1e31853f2b6ad3d16b30f2cfe71de61f2fe"],"state_sha256":"53ee9fcf5a8c74dae94976494a35953bd51b9ac1c2393fcc3087287136546e7c"}