{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:TYR6J2Y3C2ZMKRJ2BWH5S2ELKO","short_pith_number":"pith:TYR6J2Y3","schema_version":"1.0","canonical_sha256":"9e23e4eb1b16b2c5453a0d8fd9688b5386162253e6c1f39422303ff4d66147f8","source":{"kind":"arxiv","id":"2607.24331","version":1},"attestation_state":"computed","paper":{"title":"DynaCalKV: Key-Value Cache Compression via Head Grouping and Adaptive Rank Allocation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Quan V. Dang, Tan T. Nguyen","submitted_at":"2026-07-27T12:08:56Z","abstract_excerpt":"As the inference phase of Large Language Models (LLMs) requires handling long context windows, the Key-Value (KV) cache initially appears to address this challenge but eventually becomes a significant bottleneck as the context window continues to grow. Low-rank compression has recently been studied as an effective approach to reduce KV cache memory while maintaining model performance. However, only a few existing methods treat the Key and Value caches differently, despite their distinct roles. Moreover, these methods typically employ fixed attention-head grouping, which may not fully exploit t"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2607.24331","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2026-07-27T12:08:56Z","cross_cats_sorted":[],"title_canon_sha256":"b3e5e38eefc32ffd8f00756d1256cddcf5c347f70df13f22363fe008774b8021","abstract_canon_sha256":"e669310094e665ab2e1551a57865c37c2e1da7f2872cbd992f426dd9fcbd9840"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-28T02:23:57.693953Z","signature_b64":"92opfDaa2r3QJf8b3zIeuhtj/6HVoiE5hs1ow8tXdnArYhv3LDQiPGo7DicgRonCAhNIiqO2yC++OWbhfv1VCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"9e23e4eb1b16b2c5453a0d8fd9688b5386162253e6c1f39422303ff4d66147f8","last_reissued_at":"2026-07-28T02:23:57.693148Z","signature_status":"signed_v1","first_computed_at":"2026-07-28T02:23:57.693148Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"DynaCalKV: Key-Value Cache Compression via Head Grouping and Adaptive Rank Allocation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Quan V. Dang, Tan T. Nguyen","submitted_at":"2026-07-27T12:08:56Z","abstract_excerpt":"As the inference phase of Large Language Models (LLMs) requires handling long context windows, the Key-Value (KV) cache initially appears to address this challenge but eventually becomes a significant bottleneck as the context window continues to grow. Low-rank compression has recently been studied as an effective approach to reduce KV cache memory while maintaining model performance. However, only a few existing methods treat the Key and Value caches differently, despite their distinct roles. Moreover, these methods typically employ fixed attention-head grouping, which may not fully exploit t"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.24331","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/2607.24331/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"},"aliases":[{"alias_kind":"arxiv","alias_value":"2607.24331","created_at":"2026-07-28T02:23:57.693563+00:00"},{"alias_kind":"arxiv_version","alias_value":"2607.24331v1","created_at":"2026-07-28T02:23:57.693563+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.24331","created_at":"2026-07-28T02:23:57.693563+00:00"},{"alias_kind":"pith_short_12","alias_value":"TYR6J2Y3C2ZM","created_at":"2026-07-28T02:23:57.693563+00:00"},{"alias_kind":"pith_short_16","alias_value":"TYR6J2Y3C2ZMKRJ2","created_at":"2026-07-28T02:23:57.693563+00:00"},{"alias_kind":"pith_short_8","alias_value":"TYR6J2Y3","created_at":"2026-07-28T02:23:57.693563+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/TYR6J2Y3C2ZMKRJ2BWH5S2ELKO","json":"https://pith.science/pith/TYR6J2Y3C2ZMKRJ2BWH5S2ELKO.json","graph_json":"https://pith.science/api/pith-number/TYR6J2Y3C2ZMKRJ2BWH5S2ELKO/graph.json","events_json":"https://pith.science/api/pith-number/TYR6J2Y3C2ZMKRJ2BWH5S2ELKO/events.json","paper":"https://pith.science/paper/TYR6J2Y3"},"agent_actions":{"view_html":"https://pith.science/pith/TYR6J2Y3C2ZMKRJ2BWH5S2ELKO","download_json":"https://pith.science/pith/TYR6J2Y3C2ZMKRJ2BWH5S2ELKO.json","view_paper":"https://pith.science/paper/TYR6J2Y3","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2607.24331&json=true","fetch_graph":"https://pith.science/api/pith-number/TYR6J2Y3C2ZMKRJ2BWH5S2ELKO/graph.json","fetch_events":"https://pith.science/api/pith-number/TYR6J2Y3C2ZMKRJ2BWH5S2ELKO/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/TYR6J2Y3C2ZMKRJ2BWH5S2ELKO/action/timestamp_anchor","attest_storage":"https://pith.science/pith/TYR6J2Y3C2ZMKRJ2BWH5S2ELKO/action/storage_attestation","attest_author":"https://pith.science/pith/TYR6J2Y3C2ZMKRJ2BWH5S2ELKO/action/author_attestation","sign_citation":"https://pith.science/pith/TYR6J2Y3C2ZMKRJ2BWH5S2ELKO/action/citation_signature","submit_replication":"https://pith.science/pith/TYR6J2Y3C2ZMKRJ2BWH5S2ELKO/action/replication_record"}},"created_at":"2026-07-28T02:23:57.693563+00:00","updated_at":"2026-07-28T02:23:57.693563+00:00"}