{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:ZHYZRPOQFZNS2BKOTAKVERHMTY","short_pith_number":"pith:ZHYZRPOQ","schema_version":"1.0","canonical_sha256":"c9f198bdd02e5b2d054e98155244ec9e04820d49979772f226eeafb0b526b811","source":{"kind":"arxiv","id":"2606.20591","version":1},"attestation_state":"computed","paper":{"title":"Delay-Adaptive Speculation Control for Low-Latency Edge-Cloud LLM Inference","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.NI","authors_text":"Jianhua Li, Junyi He, Kangkang Sun, Minyi Guo, Xiuzhen Chen","submitted_at":"2026-05-17T07:15:30Z","abstract_excerpt":"Speculative decoding accelerates large language model (LLM) inference by using a lightweight draft model to propose tokens and a larger target model to verify them in parallel. In distributed edge-cloud inference, however, draft length must be controlled online: longer drafts amortize communication delay but reduce token acceptance, whereas shorter drafts preserve acceptance but trigger more communication rounds. We formulate this tradeoff as a ratio-type optimal stopping problem and prove that the optimal draft length is a finite delay-monotone threshold. The analysis identifies a critical de"},"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":"2606.20591","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.NI","submitted_at":"2026-05-17T07:15:30Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"b3e759e707540c75e74a91843a65b331c8c96fe4f32d26e652f7a2fe8432d464","abstract_canon_sha256":"0dc3e76857fbfb94a63cfe66c2eff3400a8d30b80deaf05e4e66fcc8989f5aa1"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-23T00:11:50.794065Z","signature_b64":"S5XQTeyR0w0CGDeNtRFGREkhml0FPvgYc5LPaJycO4zdt10hx93RnX8JkKhWndrDX8m/1X7m2aIlDG1dpitmAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c9f198bdd02e5b2d054e98155244ec9e04820d49979772f226eeafb0b526b811","last_reissued_at":"2026-06-23T00:11:50.793607Z","signature_status":"signed_v1","first_computed_at":"2026-06-23T00:11:50.793607Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Delay-Adaptive Speculation Control for Low-Latency Edge-Cloud LLM Inference","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.NI","authors_text":"Jianhua Li, Junyi He, Kangkang Sun, Minyi Guo, Xiuzhen Chen","submitted_at":"2026-05-17T07:15:30Z","abstract_excerpt":"Speculative decoding accelerates large language model (LLM) inference by using a lightweight draft model to propose tokens and a larger target model to verify them in parallel. In distributed edge-cloud inference, however, draft length must be controlled online: longer drafts amortize communication delay but reduce token acceptance, whereas shorter drafts preserve acceptance but trigger more communication rounds. We formulate this tradeoff as a ratio-type optimal stopping problem and prove that the optimal draft length is a finite delay-monotone threshold. The analysis identifies a critical de"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.20591","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/2606.20591/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":"2606.20591","created_at":"2026-06-23T00:11:50.793664+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.20591v1","created_at":"2026-06-23T00:11:50.793664+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.20591","created_at":"2026-06-23T00:11:50.793664+00:00"},{"alias_kind":"pith_short_12","alias_value":"ZHYZRPOQFZNS","created_at":"2026-06-23T00:11:50.793664+00:00"},{"alias_kind":"pith_short_16","alias_value":"ZHYZRPOQFZNS2BKO","created_at":"2026-06-23T00:11:50.793664+00:00"},{"alias_kind":"pith_short_8","alias_value":"ZHYZRPOQ","created_at":"2026-06-23T00:11:50.793664+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/ZHYZRPOQFZNS2BKOTAKVERHMTY","json":"https://pith.science/pith/ZHYZRPOQFZNS2BKOTAKVERHMTY.json","graph_json":"https://pith.science/api/pith-number/ZHYZRPOQFZNS2BKOTAKVERHMTY/graph.json","events_json":"https://pith.science/api/pith-number/ZHYZRPOQFZNS2BKOTAKVERHMTY/events.json","paper":"https://pith.science/paper/ZHYZRPOQ"},"agent_actions":{"view_html":"https://pith.science/pith/ZHYZRPOQFZNS2BKOTAKVERHMTY","download_json":"https://pith.science/pith/ZHYZRPOQFZNS2BKOTAKVERHMTY.json","view_paper":"https://pith.science/paper/ZHYZRPOQ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.20591&json=true","fetch_graph":"https://pith.science/api/pith-number/ZHYZRPOQFZNS2BKOTAKVERHMTY/graph.json","fetch_events":"https://pith.science/api/pith-number/ZHYZRPOQFZNS2BKOTAKVERHMTY/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/ZHYZRPOQFZNS2BKOTAKVERHMTY/action/timestamp_anchor","attest_storage":"https://pith.science/pith/ZHYZRPOQFZNS2BKOTAKVERHMTY/action/storage_attestation","attest_author":"https://pith.science/pith/ZHYZRPOQFZNS2BKOTAKVERHMTY/action/author_attestation","sign_citation":"https://pith.science/pith/ZHYZRPOQFZNS2BKOTAKVERHMTY/action/citation_signature","submit_replication":"https://pith.science/pith/ZHYZRPOQFZNS2BKOTAKVERHMTY/action/replication_record"}},"created_at":"2026-06-23T00:11:50.793664+00:00","updated_at":"2026-06-23T00:11:50.793664+00:00"}