{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:GZTLU2WX55X6X5HSVEB23XAKYM","short_pith_number":"pith:GZTLU2WX","canonical_record":{"source":{"id":"2505.22179","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-05-28T09:55:08Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"52e1cedaa2cee9b749f49b8a1ceb0453b1735f746d14d489dcb58902dcad4dec","abstract_canon_sha256":"16e6fb4fd0f5b69e00a5b6fbb0bbadaaa47b1f97b2c32ff730e8691bf93ad14c"},"schema_version":"1.0"},"canonical_sha256":"3666ba6ad7ef6febf4f2a903addc0ac339d3b4ade89ed107b95f6e85418513c3","source":{"kind":"arxiv","id":"2505.22179","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2505.22179","created_at":"2026-07-05T11:11:47Z"},{"alias_kind":"arxiv_version","alias_value":"2505.22179v2","created_at":"2026-07-05T11:11:47Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.22179","created_at":"2026-07-05T11:11:47Z"},{"alias_kind":"pith_short_12","alias_value":"GZTLU2WX55X6","created_at":"2026-07-05T11:11:47Z"},{"alias_kind":"pith_short_16","alias_value":"GZTLU2WX55X6X5HS","created_at":"2026-07-05T11:11:47Z"},{"alias_kind":"pith_short_8","alias_value":"GZTLU2WX","created_at":"2026-07-05T11:11:47Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:GZTLU2WX55X6X5HSVEB23XAKYM","target":"record","payload":{"canonical_record":{"source":{"id":"2505.22179","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-05-28T09:55:08Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"52e1cedaa2cee9b749f49b8a1ceb0453b1735f746d14d489dcb58902dcad4dec","abstract_canon_sha256":"16e6fb4fd0f5b69e00a5b6fbb0bbadaaa47b1f97b2c32ff730e8691bf93ad14c"},"schema_version":"1.0"},"canonical_sha256":"3666ba6ad7ef6febf4f2a903addc0ac339d3b4ade89ed107b95f6e85418513c3","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:11:47.782270Z","signature_b64":"3IhKwiqKQtyMC7ZAOlDZD2+Ygqz8LDmCtT/hbvzEvaPB83pYSCENS/e2ECL2uBT/HJVhaLXciAEQf200orf7CA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"3666ba6ad7ef6febf4f2a903addc0ac339d3b4ade89ed107b95f6e85418513c3","last_reissued_at":"2026-07-05T11:11:47.781777Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:11:47.781777Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2505.22179","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":"2026-07-05T11:11:47Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"0qT2XPDG6LIttuWtKo/P9ZlxadwLrLwoEktz+yDEAjTSMnLpBFrYUvaOFAXinxzSgbB5x1l7GQStMq8bO4gIAw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-11T11:45:46.412848Z"},"content_sha256":"1a3d7db4e3c6b64411208ee0ca9ab29df9f23262087755064d3e49f28b1f7170","schema_version":"1.0","event_id":"sha256:1a3d7db4e3c6b64411208ee0ca9ab29df9f23262087755064d3e49f28b1f7170"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:GZTLU2WX55X6X5HSVEB23XAKYM","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Speculative Decoding Meets Quantization: Compatibility Evaluation and Hierarchical Framework Design","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CL","authors_text":"Conghui Zhu, Hailong Cao, Tiejun Zhao, Wang Xu, Weilin Zhao, Xu Han, Yudi Zhang","submitted_at":"2025-05-28T09:55:08Z","abstract_excerpt":"Speculative decoding and quantization effectively accelerate memory-bound inference of large language models. Speculative decoding mitigates the memory bandwidth bottleneck by verifying multiple tokens within a single forward pass, which increases computational effort. Quantization achieves this optimization by compressing weights and activations into lower bit-widths and also reduces computations via low-bit matrix multiplications. To further leverage their strengths, we investigate the integration of these two techniques. Surprisingly, experiments applying the advanced speculative decoding m"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.22179","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/2505.22179/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":"2026-07-05T11:11:47Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"DtA/SBo3+Q9lkVbm5TMwQyZ6ZZzA4oLBVcmWcBEnZEdQaeUKvE0JyjhrnP8hUSCgFOQVjo/ixFKVuGS+DDyTCw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-11T11:45:46.413690Z"},"content_sha256":"6cc9057705877255b54de9f57d7423d40f9a0c3d9bf10448e21cd755d22d4430","schema_version":"1.0","event_id":"sha256:6cc9057705877255b54de9f57d7423d40f9a0c3d9bf10448e21cd755d22d4430"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/GZTLU2WX55X6X5HSVEB23XAKYM/bundle.json","state_url":"https://pith.science/pith/GZTLU2WX55X6X5HSVEB23XAKYM/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/GZTLU2WX55X6X5HSVEB23XAKYM/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":"2026-08-11T11:45:46Z","links":{"resolver":"https://pith.science/pith/GZTLU2WX55X6X5HSVEB23XAKYM","bundle":"https://pith.science/pith/GZTLU2WX55X6X5HSVEB23XAKYM/bundle.json","state":"https://pith.science/pith/GZTLU2WX55X6X5HSVEB23XAKYM/state.json","well_known_bundle":"https://pith.science/.well-known/pith/GZTLU2WX55X6X5HSVEB23XAKYM/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:GZTLU2WX55X6X5HSVEB23XAKYM","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":"16e6fb4fd0f5b69e00a5b6fbb0bbadaaa47b1f97b2c32ff730e8691bf93ad14c","cross_cats_sorted":["cs.AI","cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-05-28T09:55:08Z","title_canon_sha256":"52e1cedaa2cee9b749f49b8a1ceb0453b1735f746d14d489dcb58902dcad4dec"},"schema_version":"1.0","source":{"id":"2505.22179","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2505.22179","created_at":"2026-07-05T11:11:47Z"},{"alias_kind":"arxiv_version","alias_value":"2505.22179v2","created_at":"2026-07-05T11:11:47Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.22179","created_at":"2026-07-05T11:11:47Z"},{"alias_kind":"pith_short_12","alias_value":"GZTLU2WX55X6","created_at":"2026-07-05T11:11:47Z"},{"alias_kind":"pith_short_16","alias_value":"GZTLU2WX55X6X5HS","created_at":"2026-07-05T11:11:47Z"},{"alias_kind":"pith_short_8","alias_value":"GZTLU2WX","created_at":"2026-07-05T11:11:47Z"}],"graph_snapshots":[{"event_id":"sha256:6cc9057705877255b54de9f57d7423d40f9a0c3d9bf10448e21cd755d22d4430","target":"graph","created_at":"2026-07-05T11:11:47Z","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/2505.22179/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Speculative decoding and quantization effectively accelerate memory-bound inference of large language models. Speculative decoding mitigates the memory bandwidth bottleneck by verifying multiple tokens within a single forward pass, which increases computational effort. Quantization achieves this optimization by compressing weights and activations into lower bit-widths and also reduces computations via low-bit matrix multiplications. To further leverage their strengths, we investigate the integration of these two techniques. Surprisingly, experiments applying the advanced speculative decoding m","authors_text":"Conghui Zhu, Hailong Cao, Tiejun Zhao, Wang Xu, Weilin Zhao, Xu Han, Yudi Zhang","cross_cats":["cs.AI","cs.LG"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-05-28T09:55:08Z","title":"Speculative Decoding Meets Quantization: Compatibility Evaluation and Hierarchical Framework Design"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.22179","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":"sha256:1a3d7db4e3c6b64411208ee0ca9ab29df9f23262087755064d3e49f28b1f7170","target":"record","created_at":"2026-07-05T11:11:47Z","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":"16e6fb4fd0f5b69e00a5b6fbb0bbadaaa47b1f97b2c32ff730e8691bf93ad14c","cross_cats_sorted":["cs.AI","cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-05-28T09:55:08Z","title_canon_sha256":"52e1cedaa2cee9b749f49b8a1ceb0453b1735f746d14d489dcb58902dcad4dec"},"schema_version":"1.0","source":{"id":"2505.22179","kind":"arxiv","version":2}},"canonical_sha256":"3666ba6ad7ef6febf4f2a903addc0ac339d3b4ade89ed107b95f6e85418513c3","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"3666ba6ad7ef6febf4f2a903addc0ac339d3b4ade89ed107b95f6e85418513c3","first_computed_at":"2026-07-05T11:11:47.781777Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:11:47.781777Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"3IhKwiqKQtyMC7ZAOlDZD2+Ygqz8LDmCtT/hbvzEvaPB83pYSCENS/e2ECL2uBT/HJVhaLXciAEQf200orf7CA==","signature_status":"signed_v1","signed_at":"2026-07-05T11:11:47.782270Z","signed_message":"canonical_sha256_bytes"},"source_id":"2505.22179","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:1a3d7db4e3c6b64411208ee0ca9ab29df9f23262087755064d3e49f28b1f7170","sha256:6cc9057705877255b54de9f57d7423d40f9a0c3d9bf10448e21cd755d22d4430"],"state_sha256":"5544353a407ab8b283d157c415c7ff77a292f6a66502841eb24d43f449281006"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"MUu393EwXCFNsSuYw2CZMnO4RwDJrXGcHYf/Nt1fftFaWO5Jn10yACZ6cDfgzNrAODN69auV7lsPnYEvkUyCAQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-11T11:45:46.425941Z","bundle_sha256":"4abf3aee0104243e7e70e97d3af1ca274efe9d892a46f1efa844cbba8036b246"}}