{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:EEWGKQH2PA4OU37HDISKGOCNDL","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":"219ba16ef468f7d6d98bc8dd6c07845dc2ad5977599c585d9020df519514f8ef","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.DC","submitted_at":"2026-02-10T16:19:56Z","title_canon_sha256":"b2cdf3e6cae54ea497f4b57ad966171f2db553c8a8345929034a4d94a88e77c3"},"schema_version":"1.0","source":{"id":"2604.09557","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2604.09557","created_at":"2026-05-29T02:05:44Z"},{"alias_kind":"arxiv_version","alias_value":"2604.09557v2","created_at":"2026-05-29T02:05:44Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2604.09557","created_at":"2026-05-29T02:05:44Z"},{"alias_kind":"pith_short_12","alias_value":"EEWGKQH2PA4O","created_at":"2026-05-29T02:05:44Z"},{"alias_kind":"pith_short_16","alias_value":"EEWGKQH2PA4OU37H","created_at":"2026-05-29T02:05:44Z"},{"alias_kind":"pith_short_8","alias_value":"EEWGKQH2","created_at":"2026-05-29T02:05:44Z"}],"graph_snapshots":[{"event_id":"sha256:7771fc57509faa7a19a75c18df58b05477ae0fdc85be53cd9a70b9c7dab69b6e","target":"graph","created_at":"2026-05-29T02:05:44Z","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":4,"items":[{"attestation":"unclaimed","claim_id":"C1","kind":"strongest_claim","source":"verdict.strongest_claim","status":"machine_extracted","text":"SPEED-Bench establishes a unified evaluation standard for practical comparisons of SD algorithms by offering diverse semantic domains, throughput splits across concurrencies, and integration with production engines like vLLM and TensorRT-LLM."},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","text":"That the qualitative data split, curated by prioritizing semantic diversity across samples, sufficiently represents real-world workloads and that integration with the named production engines accurately exposes behaviors masked by other benchmarks."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","text":"SPEED-Bench is a new standardized benchmark for speculative decoding that supplies semantically diverse qualitative data and throughput-oriented splits across concurrency levels, integrated with vLLM and TensorRT-LLM."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"SPEED-Bench establishes a unified benchmark for speculative decoding that covers diverse semantic domains, throughput across concurrencies, and integration with production engines."}],"snapshot_sha256":"175995a3c54657284e858c8b6d83def9ca9c5dc6f75644999908b5535f472656"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2604.09557/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Speculative Decoding (SD) has emerged as a critical technique for accelerating Large Language Model (LLM) inference. Unlike deterministic system optimizations, SD performance is inherently data-dependent, meaning that diverse and representative workloads are essential for accurately measuring its effectiveness. Existing benchmarks suffer from limited task diversity, inadequate support for throughput-oriented evaluation, and a reliance on high-level implementations that fail to reflect production environments. To address this, we introduce SPEED-Bench, a comprehensive suite designed to standard","authors_text":"Benjamin Chislett, Bita Darvish Rouhani, Izzy Putterman, Maor Ashkenazi, Ran Zilberstein, Talor Abramovich, Tiyasa Mitra, Yonatan Geifman","cross_cats":["cs.AI"],"headline":"SPEED-Bench establishes a unified benchmark for speculative decoding that covers diverse semantic domains, throughput across concurrencies, and integration with production engines.","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.DC","submitted_at":"2026-02-10T16:19:56Z","title":"SPEED-Bench: A Unified and Diverse Benchmark for Speculative Decoding"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2604.09557","kind":"arxiv","version":2},"verdict":{"created_at":"2026-05-16T03:05:43.307851Z","id":"8782035f-fd2d-4952-8abc-2d3819e344e0","model_set":{"reader":"grok-4.3"},"one_line_summary":"SPEED-Bench is a new standardized benchmark for speculative decoding that supplies semantically diverse qualitative data and throughput-oriented splits across concurrency levels, integrated with vLLM and TensorRT-LLM.","pipeline_version":"pith-pipeline@v0.9.0","pith_extraction_headline":"SPEED-Bench establishes a unified benchmark for speculative decoding that covers diverse semantic domains, throughput across concurrencies, and integration with production engines.","strongest_claim":"SPEED-Bench establishes a unified evaluation standard for practical comparisons of SD algorithms by offering diverse semantic domains, throughput splits across concurrencies, and integration with production engines like vLLM and TensorRT-LLM.","weakest_assumption":"That the qualitative data split, curated by prioritizing semantic diversity across samples, sufficiently represents real-world workloads and that integration with the named production engines accurately exposes behaviors masked by other benchmarks."}},"verdict_id":"8782035f-fd2d-4952-8abc-2d3819e344e0"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:0b03da89ec75f182eecf55dafbc134a66408520ac54094897d1f964778002790","target":"record","created_at":"2026-05-29T02:05:44Z","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":"219ba16ef468f7d6d98bc8dd6c07845dc2ad5977599c585d9020df519514f8ef","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.DC","submitted_at":"2026-02-10T16:19:56Z","title_canon_sha256":"b2cdf3e6cae54ea497f4b57ad966171f2db553c8a8345929034a4d94a88e77c3"},"schema_version":"1.0","source":{"id":"2604.09557","kind":"arxiv","version":2}},"canonical_sha256":"212c6540fa7838ea6fe71a24a3384d1ad308f3be425f552f2322e3e1935205d2","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"212c6540fa7838ea6fe71a24a3384d1ad308f3be425f552f2322e3e1935205d2","first_computed_at":"2026-05-29T02:05:44.437764Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-29T02:05:44.437764Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"4pzp2P0kbXQmonuIBvGvdHXFV/VSPTiFG+fiFV9uhWxzsLr2jIo1RDMa5T1Kusggzq5caVfkftaq799d1lwbAw==","signature_status":"signed_v1","signed_at":"2026-05-29T02:05:44.438371Z","signed_message":"canonical_sha256_bytes"},"source_id":"2604.09557","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:0b03da89ec75f182eecf55dafbc134a66408520ac54094897d1f964778002790","sha256:7771fc57509faa7a19a75c18df58b05477ae0fdc85be53cd9a70b9c7dab69b6e"],"state_sha256":"a4196f4b99698ad98c72b511c7da0981f90af76794611654e2ba414fd841251a"}