{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:U2ICW3ADCJGSZAXFHAKQO3VWUN","short_pith_number":"pith:U2ICW3AD","schema_version":"1.0","canonical_sha256":"a6902b6c03124d2c82e53815076eb6a3602c4a84d4537c408cfb20f182e66c9d","source":{"kind":"arxiv","id":"2606.24957","version":1},"attestation_state":"computed","paper":{"title":"Dustin: Draft-Augmented Sparse Verification for Efficient Long-Context Generation with Speculative Decoding","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Chi-Chih Chang, Chun-Che Yang, Grace Li Zhang, Jian-Jia Chen, Kai-Chiang Wu, Ning-Chi Huang, Pei-Shuo Wang, WenHung Lee, Xiaolin Lin","submitted_at":"2026-06-23T08:51:20Z","abstract_excerpt":"While speculative decoding improves inference throughput for multi-batch long-context Large Language Models (LLMs), its efficiency is often limited by a verification bottleneck where Key-Value (KV) cache loading dominates latency. Existing compression methods fail in this regime: static eviction incurs accuracy loss due to saliency shift, while dynamic selection introduces prohibitive computational overhead during the verification path. We propose Dustin, a sparse verification framework designed for long-context speculative decoding. Dustin integrates lookahead signals from the draft model wit"},"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":"2606.24957","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-06-23T08:51:20Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"eca8282aa7557e2d9f9a9faf428ba4d4df961beef9be71b2bf87ac3be9441443","abstract_canon_sha256":"ac0f6f17fcb7e4a1eec27789304bae9542ddff0a4bdb6a3523a3c5b5d1a18094"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-25T00:17:46.879986Z","signature_b64":"MQCiz9RJu06eToRJg9ZJ5DSnBclAIkzr0ABG4Uq5uwr5iS9kK+3fB8zs/jlqriyXJoLsezbgKtNUhDNnbQNsDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"a6902b6c03124d2c82e53815076eb6a3602c4a84d4537c408cfb20f182e66c9d","last_reissued_at":"2026-06-25T00:17:46.879575Z","signature_status":"signed_v1","first_computed_at":"2026-06-25T00:17:46.879575Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Dustin: Draft-Augmented Sparse Verification for Efficient Long-Context Generation with Speculative Decoding","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Chi-Chih Chang, Chun-Che Yang, Grace Li Zhang, Jian-Jia Chen, Kai-Chiang Wu, Ning-Chi Huang, Pei-Shuo Wang, WenHung Lee, Xiaolin Lin","submitted_at":"2026-06-23T08:51:20Z","abstract_excerpt":"While speculative decoding improves inference throughput for multi-batch long-context Large Language Models (LLMs), its efficiency is often limited by a verification bottleneck where Key-Value (KV) cache loading dominates latency. Existing compression methods fail in this regime: static eviction incurs accuracy loss due to saliency shift, while dynamic selection introduces prohibitive computational overhead during the verification path. We propose Dustin, a sparse verification framework designed for long-context speculative decoding. Dustin integrates lookahead signals from the draft model wit"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.24957","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/2606.24957/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":"2606.24957","created_at":"2026-06-25T00:17:46.879640+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.24957v1","created_at":"2026-06-25T00:17:46.879640+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.24957","created_at":"2026-06-25T00:17:46.879640+00:00"},{"alias_kind":"pith_short_12","alias_value":"U2ICW3ADCJGS","created_at":"2026-06-25T00:17:46.879640+00:00"},{"alias_kind":"pith_short_16","alias_value":"U2ICW3ADCJGSZAXF","created_at":"2026-06-25T00:17:46.879640+00:00"},{"alias_kind":"pith_short_8","alias_value":"U2ICW3AD","created_at":"2026-06-25T00:17:46.879640+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/U2ICW3ADCJGSZAXFHAKQO3VWUN","json":"https://pith.science/pith/U2ICW3ADCJGSZAXFHAKQO3VWUN.json","graph_json":"https://pith.science/api/pith-number/U2ICW3ADCJGSZAXFHAKQO3VWUN/graph.json","events_json":"https://pith.science/api/pith-number/U2ICW3ADCJGSZAXFHAKQO3VWUN/events.json","paper":"https://pith.science/paper/U2ICW3AD"},"agent_actions":{"view_html":"https://pith.science/pith/U2ICW3ADCJGSZAXFHAKQO3VWUN","download_json":"https://pith.science/pith/U2ICW3ADCJGSZAXFHAKQO3VWUN.json","view_paper":"https://pith.science/paper/U2ICW3AD","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.24957&json=true","fetch_graph":"https://pith.science/api/pith-number/U2ICW3ADCJGSZAXFHAKQO3VWUN/graph.json","fetch_events":"https://pith.science/api/pith-number/U2ICW3ADCJGSZAXFHAKQO3VWUN/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/U2ICW3ADCJGSZAXFHAKQO3VWUN/action/timestamp_anchor","attest_storage":"https://pith.science/pith/U2ICW3ADCJGSZAXFHAKQO3VWUN/action/storage_attestation","attest_author":"https://pith.science/pith/U2ICW3ADCJGSZAXFHAKQO3VWUN/action/author_attestation","sign_citation":"https://pith.science/pith/U2ICW3ADCJGSZAXFHAKQO3VWUN/action/citation_signature","submit_replication":"https://pith.science/pith/U2ICW3ADCJGSZAXFHAKQO3VWUN/action/replication_record"}},"created_at":"2026-06-25T00:17:46.879640+00:00","updated_at":"2026-06-25T00:17:46.879640+00:00"}