{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:5ET53GV2L7RQ4B4FGK333GHWSC","short_pith_number":"pith:5ET53GV2","canonical_record":{"source":{"id":"2503.10325","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.DC","submitted_at":"2025-03-13T13:03:38Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"5c904f6585eae73fe08206f1cec3037193430c0f9344f8eab9eeff0f96df1745","abstract_canon_sha256":"e27f34844600bb91738fe97f964c4303bd560579be2e0b35d53d2301041426a9"},"schema_version":"1.0"},"canonical_sha256":"e927dd9aba5fe30e078532b7bd98f690808e07fe6104cf97afc1b9bce94b342f","source":{"kind":"arxiv","id":"2503.10325","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2503.10325","created_at":"2026-07-05T11:03:22Z"},{"alias_kind":"arxiv_version","alias_value":"2503.10325v2","created_at":"2026-07-05T11:03:22Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2503.10325","created_at":"2026-07-05T11:03:22Z"},{"alias_kind":"pith_short_12","alias_value":"5ET53GV2L7RQ","created_at":"2026-07-05T11:03:22Z"},{"alias_kind":"pith_short_16","alias_value":"5ET53GV2L7RQ4B4F","created_at":"2026-07-05T11:03:22Z"},{"alias_kind":"pith_short_8","alias_value":"5ET53GV2","created_at":"2026-07-05T11:03:22Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:5ET53GV2L7RQ4B4FGK333GHWSC","target":"record","payload":{"canonical_record":{"source":{"id":"2503.10325","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.DC","submitted_at":"2025-03-13T13:03:38Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"5c904f6585eae73fe08206f1cec3037193430c0f9344f8eab9eeff0f96df1745","abstract_canon_sha256":"e27f34844600bb91738fe97f964c4303bd560579be2e0b35d53d2301041426a9"},"schema_version":"1.0"},"canonical_sha256":"e927dd9aba5fe30e078532b7bd98f690808e07fe6104cf97afc1b9bce94b342f","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:03:22.126417Z","signature_b64":"MwlKYIrkOgO/klJj2pPRNsMBHU0y2gsni7CFzNJ1ttFol9vxY6IdkBrrRXojkTA3bm+J2XDDYUHMg1PbKaqsBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e927dd9aba5fe30e078532b7bd98f690808e07fe6104cf97afc1b9bce94b342f","last_reissued_at":"2026-07-05T11:03:22.125982Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:03:22.125982Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2503.10325","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":"2026-07-05T11:03:22Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"x/24dpfFm+7ri98WpOILk66+HNn17WTNVYRIt0ZcS4qjH9SCQDJkJC/yvISBmB0bjkceTgI6EoquEWCASmG9Ag==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-07T03:55:22.556349Z"},"content_sha256":"6e54c129122b7b5141ebebdfc4ad8c1948901ddd1d364c780b365722af4e2c39","schema_version":"1.0","event_id":"sha256:6e54c129122b7b5141ebebdfc4ad8c1948901ddd1d364c780b365722af4e2c39"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:5ET53GV2L7RQ4B4FGK333GHWSC","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Collaborative Speculative Inference for Efficient LLM Inference Serving","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.DC","authors_text":"Hongli Xu, Jianchun Liu, Liusheng Huang, Luyao Gao, Xichong Zhang, Yunming Liao","submitted_at":"2025-03-13T13:03:38Z","abstract_excerpt":"Speculative inference is a promising paradigm employing small speculative models (SSMs) as drafters to generate draft tokens, which are subsequently verified in parallel by the target large language model (LLM). This approach enhances the efficiency of inference serving by reducing LLM inference latency and costs while preserving generation quality. However, existing speculative methods face critical challenges, including inefficient resource utilization and limited draft acceptance, which constrain their scalability and overall effectiveness. To overcome these obstacles, we present CoSine, a "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2503.10325","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/2503.10325/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":"2026-07-05T11:03:22Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"pWN1VNUf9/c+ggHWgKO9WHU/kpECj55nVoeu9IXeK2IYzGalOfx2ZAETfgwQA2Xd8i501GW+tnGna5NHtxrFDQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-07T03:55:22.556731Z"},"content_sha256":"93aec9b62ecd18d3afe2b5c4b961dc71ab2cdd6672260b22728907ff6c5febeb","schema_version":"1.0","event_id":"sha256:93aec9b62ecd18d3afe2b5c4b961dc71ab2cdd6672260b22728907ff6c5febeb"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/5ET53GV2L7RQ4B4FGK333GHWSC/bundle.json","state_url":"https://pith.science/pith/5ET53GV2L7RQ4B4FGK333GHWSC/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/5ET53GV2L7RQ4B4FGK333GHWSC/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":"2026-07-07T03:55:22Z","links":{"resolver":"https://pith.science/pith/5ET53GV2L7RQ4B4FGK333GHWSC","bundle":"https://pith.science/pith/5ET53GV2L7RQ4B4FGK333GHWSC/bundle.json","state":"https://pith.science/pith/5ET53GV2L7RQ4B4FGK333GHWSC/state.json","well_known_bundle":"https://pith.science/.well-known/pith/5ET53GV2L7RQ4B4FGK333GHWSC/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:5ET53GV2L7RQ4B4FGK333GHWSC","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":"e27f34844600bb91738fe97f964c4303bd560579be2e0b35d53d2301041426a9","cross_cats_sorted":["cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.DC","submitted_at":"2025-03-13T13:03:38Z","title_canon_sha256":"5c904f6585eae73fe08206f1cec3037193430c0f9344f8eab9eeff0f96df1745"},"schema_version":"1.0","source":{"id":"2503.10325","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2503.10325","created_at":"2026-07-05T11:03:22Z"},{"alias_kind":"arxiv_version","alias_value":"2503.10325v2","created_at":"2026-07-05T11:03:22Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2503.10325","created_at":"2026-07-05T11:03:22Z"},{"alias_kind":"pith_short_12","alias_value":"5ET53GV2L7RQ","created_at":"2026-07-05T11:03:22Z"},{"alias_kind":"pith_short_16","alias_value":"5ET53GV2L7RQ4B4F","created_at":"2026-07-05T11:03:22Z"},{"alias_kind":"pith_short_8","alias_value":"5ET53GV2","created_at":"2026-07-05T11:03:22Z"}],"graph_snapshots":[{"event_id":"sha256:93aec9b62ecd18d3afe2b5c4b961dc71ab2cdd6672260b22728907ff6c5febeb","target":"graph","created_at":"2026-07-05T11:03:22Z","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/2503.10325/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Speculative inference is a promising paradigm employing small speculative models (SSMs) as drafters to generate draft tokens, which are subsequently verified in parallel by the target large language model (LLM). This approach enhances the efficiency of inference serving by reducing LLM inference latency and costs while preserving generation quality. However, existing speculative methods face critical challenges, including inefficient resource utilization and limited draft acceptance, which constrain their scalability and overall effectiveness. To overcome these obstacles, we present CoSine, a ","authors_text":"Hongli Xu, Jianchun Liu, Liusheng Huang, Luyao Gao, Xichong Zhang, Yunming Liao","cross_cats":["cs.LG"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.DC","submitted_at":"2025-03-13T13:03:38Z","title":"Collaborative Speculative Inference for Efficient LLM Inference Serving"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2503.10325","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":"sha256:6e54c129122b7b5141ebebdfc4ad8c1948901ddd1d364c780b365722af4e2c39","target":"record","created_at":"2026-07-05T11:03:22Z","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":"e27f34844600bb91738fe97f964c4303bd560579be2e0b35d53d2301041426a9","cross_cats_sorted":["cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.DC","submitted_at":"2025-03-13T13:03:38Z","title_canon_sha256":"5c904f6585eae73fe08206f1cec3037193430c0f9344f8eab9eeff0f96df1745"},"schema_version":"1.0","source":{"id":"2503.10325","kind":"arxiv","version":2}},"canonical_sha256":"e927dd9aba5fe30e078532b7bd98f690808e07fe6104cf97afc1b9bce94b342f","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"e927dd9aba5fe30e078532b7bd98f690808e07fe6104cf97afc1b9bce94b342f","first_computed_at":"2026-07-05T11:03:22.125982Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:03:22.125982Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"MwlKYIrkOgO/klJj2pPRNsMBHU0y2gsni7CFzNJ1ttFol9vxY6IdkBrrRXojkTA3bm+J2XDDYUHMg1PbKaqsBw==","signature_status":"signed_v1","signed_at":"2026-07-05T11:03:22.126417Z","signed_message":"canonical_sha256_bytes"},"source_id":"2503.10325","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:6e54c129122b7b5141ebebdfc4ad8c1948901ddd1d364c780b365722af4e2c39","sha256:93aec9b62ecd18d3afe2b5c4b961dc71ab2cdd6672260b22728907ff6c5febeb"],"state_sha256":"27ced3b369a85b80c8ac61328045abe36a42c8f6c1ca1b5046fb946b3a9ee096"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"pHaV16wA1JQ1GtY72SopMFc1ltzcmVsiUQLWhCgy2sYcp/bwRsgOKoA196FJex48pHstHX2J7v0KRJH4AUs7Dw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-07T03:55:22.558671Z","bundle_sha256":"ce642d3c083084f6d9abddac71b5bfe4a434f4090752f28f9d5c4bdef520704d"}}