{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:BFJLOAP54USTGBFE2FWPBJSMLD","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":"70b07c65f43d88a06f5dc469a1fe41ee99dd354264195866245f6f987557d31b","cross_cats_sorted":["cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2024-07-30T18:19:38Z","title_canon_sha256":"76a1374c7de699cc9c1887c9c00ee66aaa96826201b59d4fba9be4e0c3fb6bc3"},"schema_version":"1.0","source":{"id":"2407.21118","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2407.21118","created_at":"2026-07-05T09:30:19Z"},{"alias_kind":"arxiv_version","alias_value":"2407.21118v2","created_at":"2026-07-05T09:30:19Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2407.21118","created_at":"2026-07-05T09:30:19Z"},{"alias_kind":"pith_short_12","alias_value":"BFJLOAP54UST","created_at":"2026-07-05T09:30:19Z"},{"alias_kind":"pith_short_16","alias_value":"BFJLOAP54USTGBFE","created_at":"2026-07-05T09:30:19Z"},{"alias_kind":"pith_short_8","alias_value":"BFJLOAP5","created_at":"2026-07-05T09:30:19Z"}],"graph_snapshots":[{"event_id":"sha256:b73b8551adad6eb7cf4144b7d0a739edd9a25e3556bdfdaa7d6194c28e1038c8","target":"graph","created_at":"2026-07-05T09:30:19Z","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/2407.21118/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Post-training KV-Cache compression methods typically either sample a subset of effectual tokens or quantize the data into lower numerical bit width. However, these methods cannot exploit redundancy in the hidden dimension of the KV tensors. This paper presents a hidden dimension compression approach called Palu, a KV-Cache compression framework that utilizes low-rank projection to reduce inference-time LLM memory usage. Palu decomposes the linear layers into low-rank matrices, caches compressed intermediate states, and reconstructs the full keys and values on the fly. To improve accuracy, comp","authors_text":"Chi-Chih Chang, Chien-Yu Lin, Chong-Yan Chen, Kai-Chiang Wu, Luis Ceze, Mohamed S. Abdelfattah, Ning-Chi Huang, Pei-Shuo Wang, Wei-Cheng Lin, Yu-Fang Hu","cross_cats":["cs.LG"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2024-07-30T18:19:38Z","title":"Palu: Compressing KV-Cache with Low-Rank Projection"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2407.21118","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":"sha256:6c5b3dfb6fa05f9931263fcb4297ae200b2190ee2d6d10c0a627f9d0e1e08ef5","target":"record","created_at":"2026-07-05T09:30:19Z","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":"70b07c65f43d88a06f5dc469a1fe41ee99dd354264195866245f6f987557d31b","cross_cats_sorted":["cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2024-07-30T18:19:38Z","title_canon_sha256":"76a1374c7de699cc9c1887c9c00ee66aaa96826201b59d4fba9be4e0c3fb6bc3"},"schema_version":"1.0","source":{"id":"2407.21118","kind":"arxiv","version":2}},"canonical_sha256":"0952b701fde5253304a4d16cf0a64c58c63f4a61608d5a1fb81b670ec53b0d2e","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"0952b701fde5253304a4d16cf0a64c58c63f4a61608d5a1fb81b670ec53b0d2e","first_computed_at":"2026-07-05T09:30:19.635801Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T09:30:19.635801Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"QfP+ewPDpkO3IMgcvtEE8SwIkJ1wO+ZJLWeYEnZjEKsxe+vF2qtxSlooX3hQ3mMMMH/QTs8jKKLP/djFazxBDQ==","signature_status":"signed_v1","signed_at":"2026-07-05T09:30:19.636354Z","signed_message":"canonical_sha256_bytes"},"source_id":"2407.21118","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:6c5b3dfb6fa05f9931263fcb4297ae200b2190ee2d6d10c0a627f9d0e1e08ef5","sha256:b73b8551adad6eb7cf4144b7d0a739edd9a25e3556bdfdaa7d6194c28e1038c8"],"state_sha256":"3268553183eebcf1ce96a1c7508a0964f8dd70fef7d37dda9f02b6fe1d8c77e6"}