{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:YDKI5NWYH42AEUKSMF5JHTLGRR","short_pith_number":"pith:YDKI5NWY","canonical_record":{"source":{"id":"2410.03960","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-10-04T22:45:26Z","cross_cats_sorted":["cs.AI","cs.CL"],"title_canon_sha256":"47d429d7f94d1e557f7f1e164802aafbfeb3f23b7048c5a85d5889675045b367","abstract_canon_sha256":"b7af2443e5f140a9fa408599887bef1e3678039124d1594a60f6d1b96bb5297c"},"schema_version":"1.0"},"canonical_sha256":"c0d48eb6d83f34025152617a93cd668c6f9408d5a7b8dd0e7d7a367eeef10cf0","source":{"kind":"arxiv","id":"2410.03960","version":3},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2410.03960","created_at":"2026-07-05T11:13:58Z"},{"alias_kind":"arxiv_version","alias_value":"2410.03960v3","created_at":"2026-07-05T11:13:58Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2410.03960","created_at":"2026-07-05T11:13:58Z"},{"alias_kind":"pith_short_12","alias_value":"YDKI5NWYH42A","created_at":"2026-07-05T11:13:58Z"},{"alias_kind":"pith_short_16","alias_value":"YDKI5NWYH42AEUKS","created_at":"2026-07-05T11:13:58Z"},{"alias_kind":"pith_short_8","alias_value":"YDKI5NWY","created_at":"2026-07-05T11:13:58Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:YDKI5NWYH42AEUKSMF5JHTLGRR","target":"record","payload":{"canonical_record":{"source":{"id":"2410.03960","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-10-04T22:45:26Z","cross_cats_sorted":["cs.AI","cs.CL"],"title_canon_sha256":"47d429d7f94d1e557f7f1e164802aafbfeb3f23b7048c5a85d5889675045b367","abstract_canon_sha256":"b7af2443e5f140a9fa408599887bef1e3678039124d1594a60f6d1b96bb5297c"},"schema_version":"1.0"},"canonical_sha256":"c0d48eb6d83f34025152617a93cd668c6f9408d5a7b8dd0e7d7a367eeef10cf0","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:13:58.483438Z","signature_b64":"3jT9bcTGpxN2/g9fMt+JXpmlHjwxfPxJYrYJiM+QsL5yWy1kqzwMq/fPk0otzZ6pVRR/LCCrgBEB5NNsclY5Cw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c0d48eb6d83f34025152617a93cd668c6f9408d5a7b8dd0e7d7a367eeef10cf0","last_reissued_at":"2026-07-05T11:13:58.482884Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:13:58.482884Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2410.03960","source_version":3,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T11:13:58Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"g3vqe/GW2tH0S93Dv39cgVh0vLMdiA+m86pEgPyoPdhV989900zuSRX0STfhLLUVjlUCTcJUz+vEYdejasTUAQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-10T03:54:37.005779Z"},"content_sha256":"2a47e6a9618a8e4a7f4b05307adbc4b6149e54ca037f1ecaea23992d806b027e","schema_version":"1.0","event_id":"sha256:2a47e6a9618a8e4a7f4b05307adbc4b6149e54ca037f1ecaea23992d806b027e"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:YDKI5NWYH42AEUKSMF5JHTLGRR","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"SwiftKV: Fast Prefill-Optimized Inference with Knowledge-Preserving Model Transformation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CL"],"primary_cat":"cs.LG","authors_text":"Aurick Qiao, Samyam Rajbhandari, Yuxiong He, Zhewei Yao","submitted_at":"2024-10-04T22:45:26Z","abstract_excerpt":"LLM inference for enterprise applications, such as summarization, RAG, and code-generation, typically observe much longer prompt than generations, leading to high prefill cost and response latency. We present SwiftKV, a novel model transformation and distillation procedure targeted at reducing the prefill compute (in FLOPs) of prompt tokens while preserving high generation quality. First, SwiftKV prefills later layers' KV cache using an earlier layer's output, allowing prompt tokens to skip those later layers. Second, SwiftKV employs a lightweight knowledge-preserving distillation procedure th"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2410.03960","kind":"arxiv","version":3},"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/2410.03960/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":"2026-07-05T11:13:58Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"lkHZRhUTAVjeVx9nu4BSJLzsW24qHKtQI8mPba63/oAKQ9EAhmKUW3yxzcxo2z7QdMEXnMTIQM00JhxVa30WBA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-10T03:54:37.006390Z"},"content_sha256":"dfb71d31ee700e3066b23f50d5c9191777ad758ef51163d8759e668ad7054a2e","schema_version":"1.0","event_id":"sha256:dfb71d31ee700e3066b23f50d5c9191777ad758ef51163d8759e668ad7054a2e"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/YDKI5NWYH42AEUKSMF5JHTLGRR/bundle.json","state_url":"https://pith.science/pith/YDKI5NWYH42AEUKSMF5JHTLGRR/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/YDKI5NWYH42AEUKSMF5JHTLGRR/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":"2026-08-10T03:54:37Z","links":{"resolver":"https://pith.science/pith/YDKI5NWYH42AEUKSMF5JHTLGRR","bundle":"https://pith.science/pith/YDKI5NWYH42AEUKSMF5JHTLGRR/bundle.json","state":"https://pith.science/pith/YDKI5NWYH42AEUKSMF5JHTLGRR/state.json","well_known_bundle":"https://pith.science/.well-known/pith/YDKI5NWYH42AEUKSMF5JHTLGRR/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:YDKI5NWYH42AEUKSMF5JHTLGRR","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":"b7af2443e5f140a9fa408599887bef1e3678039124d1594a60f6d1b96bb5297c","cross_cats_sorted":["cs.AI","cs.CL"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-10-04T22:45:26Z","title_canon_sha256":"47d429d7f94d1e557f7f1e164802aafbfeb3f23b7048c5a85d5889675045b367"},"schema_version":"1.0","source":{"id":"2410.03960","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2410.03960","created_at":"2026-07-05T11:13:58Z"},{"alias_kind":"arxiv_version","alias_value":"2410.03960v3","created_at":"2026-07-05T11:13:58Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2410.03960","created_at":"2026-07-05T11:13:58Z"},{"alias_kind":"pith_short_12","alias_value":"YDKI5NWYH42A","created_at":"2026-07-05T11:13:58Z"},{"alias_kind":"pith_short_16","alias_value":"YDKI5NWYH42AEUKS","created_at":"2026-07-05T11:13:58Z"},{"alias_kind":"pith_short_8","alias_value":"YDKI5NWY","created_at":"2026-07-05T11:13:58Z"}],"graph_snapshots":[{"event_id":"sha256:dfb71d31ee700e3066b23f50d5c9191777ad758ef51163d8759e668ad7054a2e","target":"graph","created_at":"2026-07-05T11:13:58Z","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/2410.03960/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"LLM inference for enterprise applications, such as summarization, RAG, and code-generation, typically observe much longer prompt than generations, leading to high prefill cost and response latency. We present SwiftKV, a novel model transformation and distillation procedure targeted at reducing the prefill compute (in FLOPs) of prompt tokens while preserving high generation quality. First, SwiftKV prefills later layers' KV cache using an earlier layer's output, allowing prompt tokens to skip those later layers. Second, SwiftKV employs a lightweight knowledge-preserving distillation procedure th","authors_text":"Aurick Qiao, Samyam Rajbhandari, Yuxiong He, Zhewei Yao","cross_cats":["cs.AI","cs.CL"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-10-04T22:45:26Z","title":"SwiftKV: Fast Prefill-Optimized Inference with Knowledge-Preserving Model Transformation"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2410.03960","kind":"arxiv","version":3},"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":"sha256:2a47e6a9618a8e4a7f4b05307adbc4b6149e54ca037f1ecaea23992d806b027e","target":"record","created_at":"2026-07-05T11:13:58Z","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":"b7af2443e5f140a9fa408599887bef1e3678039124d1594a60f6d1b96bb5297c","cross_cats_sorted":["cs.AI","cs.CL"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-10-04T22:45:26Z","title_canon_sha256":"47d429d7f94d1e557f7f1e164802aafbfeb3f23b7048c5a85d5889675045b367"},"schema_version":"1.0","source":{"id":"2410.03960","kind":"arxiv","version":3}},"canonical_sha256":"c0d48eb6d83f34025152617a93cd668c6f9408d5a7b8dd0e7d7a367eeef10cf0","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"c0d48eb6d83f34025152617a93cd668c6f9408d5a7b8dd0e7d7a367eeef10cf0","first_computed_at":"2026-07-05T11:13:58.482884Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:13:58.482884Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"3jT9bcTGpxN2/g9fMt+JXpmlHjwxfPxJYrYJiM+QsL5yWy1kqzwMq/fPk0otzZ6pVRR/LCCrgBEB5NNsclY5Cw==","signature_status":"signed_v1","signed_at":"2026-07-05T11:13:58.483438Z","signed_message":"canonical_sha256_bytes"},"source_id":"2410.03960","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:2a47e6a9618a8e4a7f4b05307adbc4b6149e54ca037f1ecaea23992d806b027e","sha256:dfb71d31ee700e3066b23f50d5c9191777ad758ef51163d8759e668ad7054a2e"],"state_sha256":"9b532177978edeba0408d5632436da2574b3360752f689d39dc5be9908896e9c"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"8BARj5KxMEqr6W58d5iXyjToaZldwJKzW6fd9fqWftiCCn7C1/Ka30y/DbQtjKESjWQYZVyDLIQppmWaaXSDCA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-10T03:54:37.011718Z","bundle_sha256":"227bc44595a8211c07be1f480c15ed412824956c5fecc1877517d546d0c8c336"}}