{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:TWOMVSDD7GWMAFBDK3LNF3NUNA","short_pith_number":"pith:TWOMVSDD","canonical_record":{"source":{"id":"2504.04104","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-04-05T08:31:10Z","cross_cats_sorted":[],"title_canon_sha256":"7b92bc4a38f44bd5d1dd09378726aa9b1c2b47a66f0e37dd30f3b7fcbe9eb463","abstract_canon_sha256":"ad46f6cc84a181277f3b50f8c08b6d3db03c6adc164cdfd4b05f5d916b70b446"},"schema_version":"1.0"},"canonical_sha256":"9d9ccac863f9acc0142356d6d2edb468244475403182a0d382b78bad18c81872","source":{"kind":"arxiv","id":"2504.04104","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2504.04104","created_at":"2026-07-05T12:01:21Z"},{"alias_kind":"arxiv_version","alias_value":"2504.04104v2","created_at":"2026-07-05T12:01:21Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2504.04104","created_at":"2026-07-05T12:01:21Z"},{"alias_kind":"pith_short_12","alias_value":"TWOMVSDD7GWM","created_at":"2026-07-05T12:01:21Z"},{"alias_kind":"pith_short_16","alias_value":"TWOMVSDD7GWMAFBD","created_at":"2026-07-05T12:01:21Z"},{"alias_kind":"pith_short_8","alias_value":"TWOMVSDD","created_at":"2026-07-05T12:01:21Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:TWOMVSDD7GWMAFBDK3LNF3NUNA","target":"record","payload":{"canonical_record":{"source":{"id":"2504.04104","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-04-05T08:31:10Z","cross_cats_sorted":[],"title_canon_sha256":"7b92bc4a38f44bd5d1dd09378726aa9b1c2b47a66f0e37dd30f3b7fcbe9eb463","abstract_canon_sha256":"ad46f6cc84a181277f3b50f8c08b6d3db03c6adc164cdfd4b05f5d916b70b446"},"schema_version":"1.0"},"canonical_sha256":"9d9ccac863f9acc0142356d6d2edb468244475403182a0d382b78bad18c81872","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T12:01:21.763632Z","signature_b64":"pkBnMGj9XA4zFgc6nz6lvS5ITJSCbvltqeWl5QbqFz8Be8hMugaJvL2FPtnqOS3yhJGUEu+BUo8/HsfjLyTMBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"9d9ccac863f9acc0142356d6d2edb468244475403182a0d382b78bad18c81872","last_reissued_at":"2026-07-05T12:01:21.763131Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T12:01:21.763131Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2504.04104","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":"2026-07-05T12:01:21Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"gvtBG9qA4ggKkoCOOM3eXuTLNpF0h/12hiMaaVd2HIoHddtHQ+iZmGAwOlpIDxz8IIVWbbjbP2FTmrTVAvP+AQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-09T19:54:45.623980Z"},"content_sha256":"e2defc2196063d278198210e4f658af381face02cf40531e6b43f46f30e41f54","schema_version":"1.0","event_id":"sha256:e2defc2196063d278198210e4f658af381face02cf40531e6b43f46f30e41f54"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:TWOMVSDD7GWMAFBDK3LNF3NUNA","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"SpecPipe: Accelerating Pipeline Parallelism-based LLM Inference with Speculative Decoding","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Dongxiao Yu, Guanghui Zhang, Haofei Yin, Mengbai Xiao, Tinghong Li, Xiao Zhang","submitted_at":"2025-04-05T08:31:10Z","abstract_excerpt":"The demand for large language model inference is rapidly increasing. Pipeline parallelism offers a cost-effective deployment strategy for distributed inference but suffers from high service latency. While incorporating speculative decoding to pipeline parallelism improves performance, it still faces challenges of low hardware utilization and narrow speculative window. Inspired by branch prediction in instruction pipelining, we introduce SpecPipe, which fills the pipeline with speculative tokens of a request step-by-step. By maximizing the hardware utilization, SpecPipe decodes one token per pi"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2504.04104","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/2504.04104/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":"2026-07-05T12:01:21Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"SjZeO6hqB4hZaGg03Fvyzwgdqv+rwi6qCv5J35vfDMaJ6jL1ZUKcgvVbWZ5jZY6lUFjhiIoJJDWsL10FxxD/BQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-09T19:54:45.624702Z"},"content_sha256":"4d1fedb9b6e42a7913ee38034db17f89501f0d20da2e231ae94081a00c263a33","schema_version":"1.0","event_id":"sha256:4d1fedb9b6e42a7913ee38034db17f89501f0d20da2e231ae94081a00c263a33"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/TWOMVSDD7GWMAFBDK3LNF3NUNA/bundle.json","state_url":"https://pith.science/pith/TWOMVSDD7GWMAFBDK3LNF3NUNA/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/TWOMVSDD7GWMAFBDK3LNF3NUNA/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":"2026-08-09T19:54:45Z","links":{"resolver":"https://pith.science/pith/TWOMVSDD7GWMAFBDK3LNF3NUNA","bundle":"https://pith.science/pith/TWOMVSDD7GWMAFBDK3LNF3NUNA/bundle.json","state":"https://pith.science/pith/TWOMVSDD7GWMAFBDK3LNF3NUNA/state.json","well_known_bundle":"https://pith.science/.well-known/pith/TWOMVSDD7GWMAFBDK3LNF3NUNA/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:TWOMVSDD7GWMAFBDK3LNF3NUNA","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":"ad46f6cc84a181277f3b50f8c08b6d3db03c6adc164cdfd4b05f5d916b70b446","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-04-05T08:31:10Z","title_canon_sha256":"7b92bc4a38f44bd5d1dd09378726aa9b1c2b47a66f0e37dd30f3b7fcbe9eb463"},"schema_version":"1.0","source":{"id":"2504.04104","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2504.04104","created_at":"2026-07-05T12:01:21Z"},{"alias_kind":"arxiv_version","alias_value":"2504.04104v2","created_at":"2026-07-05T12:01:21Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2504.04104","created_at":"2026-07-05T12:01:21Z"},{"alias_kind":"pith_short_12","alias_value":"TWOMVSDD7GWM","created_at":"2026-07-05T12:01:21Z"},{"alias_kind":"pith_short_16","alias_value":"TWOMVSDD7GWMAFBD","created_at":"2026-07-05T12:01:21Z"},{"alias_kind":"pith_short_8","alias_value":"TWOMVSDD","created_at":"2026-07-05T12:01:21Z"}],"graph_snapshots":[{"event_id":"sha256:4d1fedb9b6e42a7913ee38034db17f89501f0d20da2e231ae94081a00c263a33","target":"graph","created_at":"2026-07-05T12:01:21Z","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/2504.04104/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"The demand for large language model inference is rapidly increasing. Pipeline parallelism offers a cost-effective deployment strategy for distributed inference but suffers from high service latency. While incorporating speculative decoding to pipeline parallelism improves performance, it still faces challenges of low hardware utilization and narrow speculative window. Inspired by branch prediction in instruction pipelining, we introduce SpecPipe, which fills the pipeline with speculative tokens of a request step-by-step. By maximizing the hardware utilization, SpecPipe decodes one token per pi","authors_text":"Dongxiao Yu, Guanghui Zhang, Haofei Yin, Mengbai Xiao, Tinghong Li, Xiao Zhang","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-04-05T08:31:10Z","title":"SpecPipe: Accelerating Pipeline Parallelism-based LLM Inference with Speculative Decoding"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2504.04104","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:e2defc2196063d278198210e4f658af381face02cf40531e6b43f46f30e41f54","target":"record","created_at":"2026-07-05T12:01:21Z","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":"ad46f6cc84a181277f3b50f8c08b6d3db03c6adc164cdfd4b05f5d916b70b446","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-04-05T08:31:10Z","title_canon_sha256":"7b92bc4a38f44bd5d1dd09378726aa9b1c2b47a66f0e37dd30f3b7fcbe9eb463"},"schema_version":"1.0","source":{"id":"2504.04104","kind":"arxiv","version":2}},"canonical_sha256":"9d9ccac863f9acc0142356d6d2edb468244475403182a0d382b78bad18c81872","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"9d9ccac863f9acc0142356d6d2edb468244475403182a0d382b78bad18c81872","first_computed_at":"2026-07-05T12:01:21.763131Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T12:01:21.763131Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"pkBnMGj9XA4zFgc6nz6lvS5ITJSCbvltqeWl5QbqFz8Be8hMugaJvL2FPtnqOS3yhJGUEu+BUo8/HsfjLyTMBw==","signature_status":"signed_v1","signed_at":"2026-07-05T12:01:21.763632Z","signed_message":"canonical_sha256_bytes"},"source_id":"2504.04104","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:e2defc2196063d278198210e4f658af381face02cf40531e6b43f46f30e41f54","sha256:4d1fedb9b6e42a7913ee38034db17f89501f0d20da2e231ae94081a00c263a33"],"state_sha256":"dc3e0144b52ca8fd0a5fdb5b26dd1dce02dbda2162d8128b261ba7c1e93dd78a"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"HqUTSORvXpiZ7cODs9YsTa8IryqyUfbw5O7iwS3+PHcwpUMZMjm90EvmiHsu+TgLBNgwoF4KqWKPEfBIE+3ZAA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-09T19:54:45.631713Z","bundle_sha256":"44513afbde4119f5f5ded2d1033512201c79cce95eccf279b9721e675976b459"}}