{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:CKU6PBOFGPEDETMQNEC3KI5UAD","short_pith_number":"pith:CKU6PBOF","canonical_record":{"source":{"id":"2505.19586","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-05-26T07:00:04Z","cross_cats_sorted":[],"title_canon_sha256":"096626648fb6d7b18b885249e5d6b56d042115c5f706a6566d53a2cf2aef0240","abstract_canon_sha256":"1ceb1fd2aba448d6df7a03641bb585caa5f14527422dc7e4b91e94e4ac8932b5"},"schema_version":"1.0"},"canonical_sha256":"12a9e785c533c8324d906905b523b400e01774f0e020db4b62365a74e860dba1","source":{"kind":"arxiv","id":"2505.19586","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2505.19586","created_at":"2026-07-05T11:10:10Z"},{"alias_kind":"arxiv_version","alias_value":"2505.19586v2","created_at":"2026-07-05T11:10:10Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.19586","created_at":"2026-07-05T11:10:10Z"},{"alias_kind":"pith_short_12","alias_value":"CKU6PBOFGPED","created_at":"2026-07-05T11:10:10Z"},{"alias_kind":"pith_short_16","alias_value":"CKU6PBOFGPEDETMQ","created_at":"2026-07-05T11:10:10Z"},{"alias_kind":"pith_short_8","alias_value":"CKU6PBOF","created_at":"2026-07-05T11:10:10Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:CKU6PBOFGPEDETMQNEC3KI5UAD","target":"record","payload":{"canonical_record":{"source":{"id":"2505.19586","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-05-26T07:00:04Z","cross_cats_sorted":[],"title_canon_sha256":"096626648fb6d7b18b885249e5d6b56d042115c5f706a6566d53a2cf2aef0240","abstract_canon_sha256":"1ceb1fd2aba448d6df7a03641bb585caa5f14527422dc7e4b91e94e4ac8932b5"},"schema_version":"1.0"},"canonical_sha256":"12a9e785c533c8324d906905b523b400e01774f0e020db4b62365a74e860dba1","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:10:10.130766Z","signature_b64":"HuwL3XBUqVbltQ4Po9MZq2lGCL4ei4phnM6VwWODEz0+H4LRNY5n3ujvZpLYA5uKdw5UCNQAErtPUhQLXY0lDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"12a9e785c533c8324d906905b523b400e01774f0e020db4b62365a74e860dba1","last_reissued_at":"2026-07-05T11:10:10.130291Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:10:10.130291Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2505.19586","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T11:10:10Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"OpAZ7Dm9KgZ+2n+34U7E46qTpKYkFe4W2Vg/+HlYnG6ZTf/iD3aDHN3rZpYqDhbfeN96JuY8H1agzrx5PthpAA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-07T21:20:52.342920Z"},"content_sha256":"ab788d99d8dc7471f619ff13313b8779c49a5afe7e0ee91243b654b8f5df7ca6","schema_version":"1.0","event_id":"sha256:ab788d99d8dc7471f619ff13313b8779c49a5afe7e0ee91243b654b8f5df7ca6"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:CKU6PBOFGPEDETMQNEC3KI5UAD","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"TailorKV: A Hybrid Framework for Long-Context Inference via Tailored KV Cache Optimization","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Bin Wang, Bowen Shen, Dingyu Yao, Jian Luan, Wei Liu, Weiping Wang, Zheng Lin","submitted_at":"2025-05-26T07:00:04Z","abstract_excerpt":"The Key-Value (KV) cache in generative large language models (LLMs) introduces substantial memory overhead. Existing works mitigate this burden by offloading or compressing the KV cache. However, loading the entire cache incurs significant latency due to PCIe bandwidth bottlenecks in CPU-GPU communication, while aggressive compression causes notable performance degradation. We identify that certain layers in the LLM need to maintain global information and are unsuitable for selective loading. In contrast, other layers primarily focus on a few tokens with dominant activations that potentially i"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.19586","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/2505.19586/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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T11:10:10Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"wThjfyIrPEv9gKSpsqaid4zYOpg0tIFI3hwYSDdXWlfCcA8krqCVFUtesWuTYmb4z2a7maNctkPQ5+Sc/vcoDQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-07T21:20:52.343487Z"},"content_sha256":"7134cca2a24b0778e8432c0df4c02be1c9683f6bfc04c909d541db03147c385d","schema_version":"1.0","event_id":"sha256:7134cca2a24b0778e8432c0df4c02be1c9683f6bfc04c909d541db03147c385d"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/CKU6PBOFGPEDETMQNEC3KI5UAD/bundle.json","state_url":"https://pith.science/pith/CKU6PBOFGPEDETMQNEC3KI5UAD/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/CKU6PBOFGPEDETMQNEC3KI5UAD/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-08-07T21:20:52Z","links":{"resolver":"https://pith.science/pith/CKU6PBOFGPEDETMQNEC3KI5UAD","bundle":"https://pith.science/pith/CKU6PBOFGPEDETMQNEC3KI5UAD/bundle.json","state":"https://pith.science/pith/CKU6PBOFGPEDETMQNEC3KI5UAD/state.json","well_known_bundle":"https://pith.science/.well-known/pith/CKU6PBOFGPEDETMQNEC3KI5UAD/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:CKU6PBOFGPEDETMQNEC3KI5UAD","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":"1ceb1fd2aba448d6df7a03641bb585caa5f14527422dc7e4b91e94e4ac8932b5","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-05-26T07:00:04Z","title_canon_sha256":"096626648fb6d7b18b885249e5d6b56d042115c5f706a6566d53a2cf2aef0240"},"schema_version":"1.0","source":{"id":"2505.19586","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2505.19586","created_at":"2026-07-05T11:10:10Z"},{"alias_kind":"arxiv_version","alias_value":"2505.19586v2","created_at":"2026-07-05T11:10:10Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.19586","created_at":"2026-07-05T11:10:10Z"},{"alias_kind":"pith_short_12","alias_value":"CKU6PBOFGPED","created_at":"2026-07-05T11:10:10Z"},{"alias_kind":"pith_short_16","alias_value":"CKU6PBOFGPEDETMQ","created_at":"2026-07-05T11:10:10Z"},{"alias_kind":"pith_short_8","alias_value":"CKU6PBOF","created_at":"2026-07-05T11:10:10Z"}],"graph_snapshots":[{"event_id":"sha256:7134cca2a24b0778e8432c0df4c02be1c9683f6bfc04c909d541db03147c385d","target":"graph","created_at":"2026-07-05T11:10:10Z","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/2505.19586/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"The Key-Value (KV) cache in generative large language models (LLMs) introduces substantial memory overhead. Existing works mitigate this burden by offloading or compressing the KV cache. However, loading the entire cache incurs significant latency due to PCIe bandwidth bottlenecks in CPU-GPU communication, while aggressive compression causes notable performance degradation. We identify that certain layers in the LLM need to maintain global information and are unsuitable for selective loading. In contrast, other layers primarily focus on a few tokens with dominant activations that potentially i","authors_text":"Bin Wang, Bowen Shen, Dingyu Yao, Jian Luan, Wei Liu, Weiping Wang, Zheng Lin","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-05-26T07:00:04Z","title":"TailorKV: A Hybrid Framework for Long-Context Inference via Tailored KV Cache Optimization"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.19586","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:ab788d99d8dc7471f619ff13313b8779c49a5afe7e0ee91243b654b8f5df7ca6","target":"record","created_at":"2026-07-05T11:10:10Z","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":"1ceb1fd2aba448d6df7a03641bb585caa5f14527422dc7e4b91e94e4ac8932b5","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-05-26T07:00:04Z","title_canon_sha256":"096626648fb6d7b18b885249e5d6b56d042115c5f706a6566d53a2cf2aef0240"},"schema_version":"1.0","source":{"id":"2505.19586","kind":"arxiv","version":2}},"canonical_sha256":"12a9e785c533c8324d906905b523b400e01774f0e020db4b62365a74e860dba1","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"12a9e785c533c8324d906905b523b400e01774f0e020db4b62365a74e860dba1","first_computed_at":"2026-07-05T11:10:10.130291Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:10:10.130291Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"HuwL3XBUqVbltQ4Po9MZq2lGCL4ei4phnM6VwWODEz0+H4LRNY5n3ujvZpLYA5uKdw5UCNQAErtPUhQLXY0lDQ==","signature_status":"signed_v1","signed_at":"2026-07-05T11:10:10.130766Z","signed_message":"canonical_sha256_bytes"},"source_id":"2505.19586","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:ab788d99d8dc7471f619ff13313b8779c49a5afe7e0ee91243b654b8f5df7ca6","sha256:7134cca2a24b0778e8432c0df4c02be1c9683f6bfc04c909d541db03147c385d"],"state_sha256":"b8c26783fe68ea055f47c571e09f2fa7bc74472dfc554fefaff410482842be70"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"7OPEYUC6GRUpMDPTwpfBIgyVCnkGyXQl56zHWSlwGylWlO/dDAP9eVzQgZGGv8B/hB7batllQoIel9zxuBR5BQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-07T21:20:52.347272Z","bundle_sha256":"5949fbd47793d6c9768b9c651b2f0abd4ee9def1136934e71b88bb1ae9cab4fa"}}