{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:ISEHAEA4HAMM3VET7MNGJ2D2II","short_pith_number":"pith:ISEHAEA4","schema_version":"1.0","canonical_sha256":"448870101c3818cdd493fb1a64e87a42160f742d73f5cf6b9b6b2e1b39633922","source":{"kind":"arxiv","id":"2407.01527","version":2},"attestation_state":"computed","paper":{"title":"KV Cache Compression, But What Must We Give in Return? A Comprehensive Benchmark of Long Context Capable Approaches","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Duy Le, Guanchu Wang, Hongye Jin, Hongyi Liu, Jiayi Yuan, Shaochen Zhong, Songchen Li, Vipin Chaudhary, Xia Hu, Yu-Neng Chuang, Zhaozhuo Xu, Zirui Liu","submitted_at":"2024-07-01T17:59:47Z","abstract_excerpt":"Long context capability is a crucial competency for large language models (LLMs) as it mitigates the human struggle to digest long-form texts. This capability enables complex task-solving scenarios such as book summarization, code assistance, and many more tasks that are traditionally manpower-intensive. However, transformer-based LLMs face significant challenges with long context input due to the growing size of the KV cache and the intrinsic complexity of attending to extended inputs; where multiple schools of efficiency-driven approaches - such as KV cache quantization, token dropping, prom"},"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":"2407.01527","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-07-01T17:59:47Z","cross_cats_sorted":[],"title_canon_sha256":"9727d86c4f9d9b2124124fa80ce3726336f012078100566689263ebcfe232147","abstract_canon_sha256":"c5795ad51a12c00d0ca9bcf637d7c106bfb16582c4aab77665e8d2538102356f"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:17:44.932794Z","signature_b64":"qAhLmC9R2Zq5Olx3QaYGuMSBV+x7q+d2/gXTwJB2oHGXX6y9zulocPdz3XhtoECHI6dFR2vPqDOUbYDBK5vVCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"448870101c3818cdd493fb1a64e87a42160f742d73f5cf6b9b6b2e1b39633922","last_reissued_at":"2026-07-05T09:17:44.932294Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:17:44.932294Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"KV Cache Compression, But What Must We Give in Return? A Comprehensive Benchmark of Long Context Capable Approaches","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Duy Le, Guanchu Wang, Hongye Jin, Hongyi Liu, Jiayi Yuan, Shaochen Zhong, Songchen Li, Vipin Chaudhary, Xia Hu, Yu-Neng Chuang, Zhaozhuo Xu, Zirui Liu","submitted_at":"2024-07-01T17:59:47Z","abstract_excerpt":"Long context capability is a crucial competency for large language models (LLMs) as it mitigates the human struggle to digest long-form texts. This capability enables complex task-solving scenarios such as book summarization, code assistance, and many more tasks that are traditionally manpower-intensive. However, transformer-based LLMs face significant challenges with long context input due to the growing size of the KV cache and the intrinsic complexity of attending to extended inputs; where multiple schools of efficiency-driven approaches - such as KV cache quantization, token dropping, prom"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2407.01527","kind":"arxiv","version":2},"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/2407.01527/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":"2407.01527","created_at":"2026-07-05T09:17:44.932363+00:00"},{"alias_kind":"arxiv_version","alias_value":"2407.01527v2","created_at":"2026-07-05T09:17:44.932363+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2407.01527","created_at":"2026-07-05T09:17:44.932363+00:00"},{"alias_kind":"pith_short_12","alias_value":"ISEHAEA4HAMM","created_at":"2026-07-05T09:17:44.932363+00:00"},{"alias_kind":"pith_short_16","alias_value":"ISEHAEA4HAMM3VET","created_at":"2026-07-05T09:17:44.932363+00:00"},{"alias_kind":"pith_short_8","alias_value":"ISEHAEA4","created_at":"2026-07-05T09:17:44.932363+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":5,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.28349","citing_title":"HMARS: A Hierarchical Multi-Agent Memory System for Long-Context Reasoning","ref_index":18,"is_internal_anchor":false},{"citing_arxiv_id":"2410.13846","citing_title":"LightTransfer: Your Long-Context LLM is Secretly a Hybrid Model with Effortless Adaptation","ref_index":37,"is_internal_anchor":false},{"citing_arxiv_id":"2502.01941","citing_title":"Semantic Integrity Matters: Benchmarking and Preserving High-Density Reasoning in KV Cache Compression","ref_index":71,"is_internal_anchor":false},{"citing_arxiv_id":"2402.02750","citing_title":"KIVI: A Tuning-Free Asymmetric 2bit Quantization for KV Cache","ref_index":20,"is_internal_anchor":false},{"citing_arxiv_id":"2604.18567","citing_title":"Latent Phase-Shift Rollback: Inference-Time Error Correction via Residual Stream Monitoring and KV-Cache Steering","ref_index":22,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/ISEHAEA4HAMM3VET7MNGJ2D2II","json":"https://pith.science/pith/ISEHAEA4HAMM3VET7MNGJ2D2II.json","graph_json":"https://pith.science/api/pith-number/ISEHAEA4HAMM3VET7MNGJ2D2II/graph.json","events_json":"https://pith.science/api/pith-number/ISEHAEA4HAMM3VET7MNGJ2D2II/events.json","paper":"https://pith.science/paper/ISEHAEA4"},"agent_actions":{"view_html":"https://pith.science/pith/ISEHAEA4HAMM3VET7MNGJ2D2II","download_json":"https://pith.science/pith/ISEHAEA4HAMM3VET7MNGJ2D2II.json","view_paper":"https://pith.science/paper/ISEHAEA4","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2407.01527&json=true","fetch_graph":"https://pith.science/api/pith-number/ISEHAEA4HAMM3VET7MNGJ2D2II/graph.json","fetch_events":"https://pith.science/api/pith-number/ISEHAEA4HAMM3VET7MNGJ2D2II/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/ISEHAEA4HAMM3VET7MNGJ2D2II/action/timestamp_anchor","attest_storage":"https://pith.science/pith/ISEHAEA4HAMM3VET7MNGJ2D2II/action/storage_attestation","attest_author":"https://pith.science/pith/ISEHAEA4HAMM3VET7MNGJ2D2II/action/author_attestation","sign_citation":"https://pith.science/pith/ISEHAEA4HAMM3VET7MNGJ2D2II/action/citation_signature","submit_replication":"https://pith.science/pith/ISEHAEA4HAMM3VET7MNGJ2D2II/action/replication_record"}},"created_at":"2026-07-05T09:17:44.932363+00:00","updated_at":"2026-07-05T09:17:44.932363+00:00"}