{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:UYTKMATY3SKXOZNAKYSNODBVBZ","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":"d79ad356494da579682964a7c9e133ef518c7a11e5f8d23145dcdd446b813ebe","cross_cats_sorted":["cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-03-14T23:40:56Z","title_canon_sha256":"923a728475e0926bacfb7d17c8e0a75cdfc1205dcd3fc05fee858c505068a673"},"schema_version":"1.0","source":{"id":"2403.09919","kind":"arxiv","version":5}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2403.09919","created_at":"2026-07-05T09:49:03Z"},{"alias_kind":"arxiv_version","alias_value":"2403.09919v5","created_at":"2026-07-05T09:49:03Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2403.09919","created_at":"2026-07-05T09:49:03Z"},{"alias_kind":"pith_short_12","alias_value":"UYTKMATY3SKX","created_at":"2026-07-05T09:49:03Z"},{"alias_kind":"pith_short_16","alias_value":"UYTKMATY3SKXOZNA","created_at":"2026-07-05T09:49:03Z"},{"alias_kind":"pith_short_8","alias_value":"UYTKMATY","created_at":"2026-07-05T09:49:03Z"}],"graph_snapshots":[{"event_id":"sha256:55bd7196c7c8f6efb2fc7d6b65044541b5fdd7381474b3b7f8f6b2d98d9a6ff7","target":"graph","created_at":"2026-07-05T09:49:03Z","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/2403.09919/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"We present Recurrent Drafter (ReDrafter), an advanced speculative decoding approach that achieves state-of-the-art speedup for large language models (LLMs) inference. The performance gains are driven by three key aspects: (1) leveraging a recurrent neural network (RNN) as the draft model conditioning on LLM's hidden states, (2) applying a dynamic tree attention algorithm over beam search results to eliminate duplicated prefixes in candidate sequences, and (3) training through knowledge distillation from the LLM. ReDrafter accelerates Vicuna inference in MT-Bench by up to 2.8x with a PyTorch im","authors_text":"Aonan Zhang, Chong Wang, Xuanyu Zhang, Yi Wang, Yunfei Cheng","cross_cats":["cs.LG"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-03-14T23:40:56Z","title":"Recurrent Drafter for Fast Speculative Decoding in Large Language Models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2403.09919","kind":"arxiv","version":5},"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":"sha256:5f0733faf776d9dbfd09a34deeebf2d63bdc74a474953a45e98458bd5657be66","target":"record","created_at":"2026-07-05T09:49:03Z","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":"d79ad356494da579682964a7c9e133ef518c7a11e5f8d23145dcdd446b813ebe","cross_cats_sorted":["cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-03-14T23:40:56Z","title_canon_sha256":"923a728475e0926bacfb7d17c8e0a75cdfc1205dcd3fc05fee858c505068a673"},"schema_version":"1.0","source":{"id":"2403.09919","kind":"arxiv","version":5}},"canonical_sha256":"a626a60278dc957765a05624d70c350e5357dbe627d1a04b27c095afd3ff3515","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"a626a60278dc957765a05624d70c350e5357dbe627d1a04b27c095afd3ff3515","first_computed_at":"2026-07-05T09:49:03.437328Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T09:49:03.437328Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"PUUl+KanIprg+d//Fk6ZTICG2pwEAZcQDeTLDdtzzNCZq+k6SjjdQFPUSDjyz5w/dx8+BEzQdPsr0c8Y/YT5BQ==","signature_status":"signed_v1","signed_at":"2026-07-05T09:49:03.437839Z","signed_message":"canonical_sha256_bytes"},"source_id":"2403.09919","source_kind":"arxiv","source_version":5}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:5f0733faf776d9dbfd09a34deeebf2d63bdc74a474953a45e98458bd5657be66","sha256:55bd7196c7c8f6efb2fc7d6b65044541b5fdd7381474b3b7f8f6b2d98d9a6ff7"],"state_sha256":"560d4ccefec965f8dc32525ff46c4138136054613c08f2c69fddb9d35b2e4c82"}