{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:GUQVAWOQ2ZGBS43GZ2S2H43F3J","short_pith_number":"pith:GUQVAWOQ","schema_version":"1.0","canonical_sha256":"35215059d0d64c197366cea5a3f365da651f810328984541bac55bf07710bb21","source":{"kind":"arxiv","id":"2605.30852","version":1},"attestation_state":"computed","paper":{"title":"Speculative Pipeline Decoding: Higher-Accruacy and Zero-Bubble Speculation via Pipeline Parallelism","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Huazheng Wang, Ji Pei, Ruilong Ren, Shuai Yuan, Yijiong Yu","submitted_at":"2026-05-29T05:17:03Z","abstract_excerpt":"Speculative Decoding (SD) accelerates low-concurrency LLM inference by employing a draft-then-verify paradigm. However, mainstream methods typically rely on multi-token prediction, which introduces escalating prediction difficulty and serial drafting latency. To address these, we propose Speculative Pipeline Decoding (SPD), a groundbreaking framework that unlocks the true potential of pipeline parallelism. By partitioning the target LLM into $n$ pipeline stages, SPD allows LLM to process $n$ tokens in parallel to accelerate decoding. To continuous fill the pipeline in single sequence decoding,"},"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":"2605.30852","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2026-05-29T05:17:03Z","cross_cats_sorted":[],"title_canon_sha256":"2c975c6e86c6dcd1632cbdafbaa26382d1170f2315a102ce5eea63028cc3eb16","abstract_canon_sha256":"dd5d3ed4665e4d7f2d3dc4c6dd55aedef4f1dffca692a888dee8da4e993265b0"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-01T01:03:21.072033Z","signature_b64":"Z0lXmPNF1qj8RC9wTlmvG9ekRtbcOgaSkShDFXOw/5pMfa3LJbEh0vkRMzZVUPoAoehZX2lDXcdCwrD4uS1JAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"35215059d0d64c197366cea5a3f365da651f810328984541bac55bf07710bb21","last_reissued_at":"2026-06-01T01:03:21.071455Z","signature_status":"signed_v1","first_computed_at":"2026-06-01T01:03:21.071455Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Speculative Pipeline Decoding: Higher-Accruacy and Zero-Bubble Speculation via Pipeline Parallelism","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Huazheng Wang, Ji Pei, Ruilong Ren, Shuai Yuan, Yijiong Yu","submitted_at":"2026-05-29T05:17:03Z","abstract_excerpt":"Speculative Decoding (SD) accelerates low-concurrency LLM inference by employing a draft-then-verify paradigm. However, mainstream methods typically rely on multi-token prediction, which introduces escalating prediction difficulty and serial drafting latency. To address these, we propose Speculative Pipeline Decoding (SPD), a groundbreaking framework that unlocks the true potential of pipeline parallelism. By partitioning the target LLM into $n$ pipeline stages, SPD allows LLM to process $n$ tokens in parallel to accelerate decoding. To continuous fill the pipeline in single sequence decoding,"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.30852","kind":"arxiv","version":1},"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/2605.30852/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":"2605.30852","created_at":"2026-06-01T01:03:21.071535+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.30852v1","created_at":"2026-06-01T01:03:21.071535+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.30852","created_at":"2026-06-01T01:03:21.071535+00:00"},{"alias_kind":"pith_short_12","alias_value":"GUQVAWOQ2ZGB","created_at":"2026-06-01T01:03:21.071535+00:00"},{"alias_kind":"pith_short_16","alias_value":"GUQVAWOQ2ZGBS43G","created_at":"2026-06-01T01:03:21.071535+00:00"},{"alias_kind":"pith_short_8","alias_value":"GUQVAWOQ","created_at":"2026-06-01T01:03:21.071535+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/GUQVAWOQ2ZGBS43GZ2S2H43F3J","json":"https://pith.science/pith/GUQVAWOQ2ZGBS43GZ2S2H43F3J.json","graph_json":"https://pith.science/api/pith-number/GUQVAWOQ2ZGBS43GZ2S2H43F3J/graph.json","events_json":"https://pith.science/api/pith-number/GUQVAWOQ2ZGBS43GZ2S2H43F3J/events.json","paper":"https://pith.science/paper/GUQVAWOQ"},"agent_actions":{"view_html":"https://pith.science/pith/GUQVAWOQ2ZGBS43GZ2S2H43F3J","download_json":"https://pith.science/pith/GUQVAWOQ2ZGBS43GZ2S2H43F3J.json","view_paper":"https://pith.science/paper/GUQVAWOQ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.30852&json=true","fetch_graph":"https://pith.science/api/pith-number/GUQVAWOQ2ZGBS43GZ2S2H43F3J/graph.json","fetch_events":"https://pith.science/api/pith-number/GUQVAWOQ2ZGBS43GZ2S2H43F3J/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/GUQVAWOQ2ZGBS43GZ2S2H43F3J/action/timestamp_anchor","attest_storage":"https://pith.science/pith/GUQVAWOQ2ZGBS43GZ2S2H43F3J/action/storage_attestation","attest_author":"https://pith.science/pith/GUQVAWOQ2ZGBS43GZ2S2H43F3J/action/author_attestation","sign_citation":"https://pith.science/pith/GUQVAWOQ2ZGBS43GZ2S2H43F3J/action/citation_signature","submit_replication":"https://pith.science/pith/GUQVAWOQ2ZGBS43GZ2S2H43F3J/action/replication_record"}},"created_at":"2026-06-01T01:03:21.071535+00:00","updated_at":"2026-06-01T01:03:21.071535+00:00"}