{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:I4S7SYULJFNG5RVMQ6HIOP2CR6","short_pith_number":"pith:I4S7SYUL","schema_version":"1.0","canonical_sha256":"4725f9628b495a6ec6ac878e873f428f8aa6ff223d80e2c6cea4ae01585a4385","source":{"kind":"arxiv","id":"2505.08600","version":1},"attestation_state":"computed","paper":{"title":"Automatic Task Detection and Heterogeneous LLM Speculative Decoding","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Danying Ge, Jianhua Gao, Qizhi Jiang, Weixing Ji, Yifei Feng","submitted_at":"2025-05-13T14:16:12Z","abstract_excerpt":"Speculative decoding, which combines a draft model with a target model, has emerged as an effective approach to accelerate large language model (LLM) inference. However, existing methods often face a trade-off between the acceptance rate and decoding speed in downstream tasks due to the limited capacity of the draft model, making it difficult to ensure efficiency across diverse tasks. To address this problem, we propose a speculative decoding algorithm tailored for downstream task optimization. It includes an automatic task partitioning and assigning method, which automatically categorizes dow"},"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":"2505.08600","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-05-13T14:16:12Z","cross_cats_sorted":[],"title_canon_sha256":"eba1fc5f3cf1b72911494d045f5d8c9e8d520f8d51068d5cdef2950dd919d979","abstract_canon_sha256":"43d2d5bb58e405adc3db3b48b7be3e750f0033cdbf1f0275be3b4af1af812bf3"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:02:33.339384Z","signature_b64":"URwyXlgdDfdeBsDDGSMH4dr+jEx2aVWTxc2hXcgYoWxQNA2XJmQrjmtKqazX29lbNNK1K8y+XsHL7sqgA8CGCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"4725f9628b495a6ec6ac878e873f428f8aa6ff223d80e2c6cea4ae01585a4385","last_reissued_at":"2026-07-05T11:02:33.338889Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:02:33.338889Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Automatic Task Detection and Heterogeneous LLM Speculative Decoding","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Danying Ge, Jianhua Gao, Qizhi Jiang, Weixing Ji, Yifei Feng","submitted_at":"2025-05-13T14:16:12Z","abstract_excerpt":"Speculative decoding, which combines a draft model with a target model, has emerged as an effective approach to accelerate large language model (LLM) inference. However, existing methods often face a trade-off between the acceptance rate and decoding speed in downstream tasks due to the limited capacity of the draft model, making it difficult to ensure efficiency across diverse tasks. To address this problem, we propose a speculative decoding algorithm tailored for downstream task optimization. It includes an automatic task partitioning and assigning method, which automatically categorizes dow"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.08600","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/2505.08600/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":"2505.08600","created_at":"2026-07-05T11:02:33.338949+00:00"},{"alias_kind":"arxiv_version","alias_value":"2505.08600v1","created_at":"2026-07-05T11:02:33.338949+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.08600","created_at":"2026-07-05T11:02:33.338949+00:00"},{"alias_kind":"pith_short_12","alias_value":"I4S7SYULJFNG","created_at":"2026-07-05T11:02:33.338949+00:00"},{"alias_kind":"pith_short_16","alias_value":"I4S7SYULJFNG5RVM","created_at":"2026-07-05T11:02:33.338949+00:00"},{"alias_kind":"pith_short_8","alias_value":"I4S7SYUL","created_at":"2026-07-05T11:02:33.338949+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/I4S7SYULJFNG5RVMQ6HIOP2CR6","json":"https://pith.science/pith/I4S7SYULJFNG5RVMQ6HIOP2CR6.json","graph_json":"https://pith.science/api/pith-number/I4S7SYULJFNG5RVMQ6HIOP2CR6/graph.json","events_json":"https://pith.science/api/pith-number/I4S7SYULJFNG5RVMQ6HIOP2CR6/events.json","paper":"https://pith.science/paper/I4S7SYUL"},"agent_actions":{"view_html":"https://pith.science/pith/I4S7SYULJFNG5RVMQ6HIOP2CR6","download_json":"https://pith.science/pith/I4S7SYULJFNG5RVMQ6HIOP2CR6.json","view_paper":"https://pith.science/paper/I4S7SYUL","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2505.08600&json=true","fetch_graph":"https://pith.science/api/pith-number/I4S7SYULJFNG5RVMQ6HIOP2CR6/graph.json","fetch_events":"https://pith.science/api/pith-number/I4S7SYULJFNG5RVMQ6HIOP2CR6/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/I4S7SYULJFNG5RVMQ6HIOP2CR6/action/timestamp_anchor","attest_storage":"https://pith.science/pith/I4S7SYULJFNG5RVMQ6HIOP2CR6/action/storage_attestation","attest_author":"https://pith.science/pith/I4S7SYULJFNG5RVMQ6HIOP2CR6/action/author_attestation","sign_citation":"https://pith.science/pith/I4S7SYULJFNG5RVMQ6HIOP2CR6/action/citation_signature","submit_replication":"https://pith.science/pith/I4S7SYULJFNG5RVMQ6HIOP2CR6/action/replication_record"}},"created_at":"2026-07-05T11:02:33.338949+00:00","updated_at":"2026-07-05T11:02:33.338949+00:00"}