{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:3HSDAIB7UVULKHJ7OZZFLORDN2","short_pith_number":"pith:3HSDAIB7","schema_version":"1.0","canonical_sha256":"d9e430203fa568b51d3f767255ba236e9ef26fd536bbb77f679b7b5d074f7ab7","source":{"kind":"arxiv","id":"2402.12280","version":2},"attestation_state":"computed","paper":{"title":"Plato: Plan to Efficiently Decode for Large Language Model Inference","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Atul Prakash, Danyang Zhuo, Feng Qian, Haizhong Zheng, Matthew Lentz, Qingzhao Zhang, Shuowei Jin, Xueshen Liu, Yongji Wu, Z. Morley Mao","submitted_at":"2024-02-19T16:47:04Z","abstract_excerpt":"Large language models (LLMs) have achieved remarkable success in natural language tasks, but their inference incurs substantial computational and memory overhead. To improve efficiency, parallel decoding methods like Skeleton-of-Thought (SoT) decompose prompts into sub-problems for concurrent processing. However, these methods significantly compromise answer quality by treating semantically linked sub-problems as independent. We propose Plato, a novel approach that co-designs algorithms and systems for semantic-aware parallel decoding. Plato leverages LLMs to organize sub-problems into a depen"},"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":"2402.12280","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-02-19T16:47:04Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"47709556b977ee482a167d4f2d63ab2a04cf048e2f85ae1c812e485397dbd9ff","abstract_canon_sha256":"9e7325209d88c2dc6482819d16d44c54c37e65dce34b48aaa27f564f144714bb"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:48:14.690565Z","signature_b64":"Hse/X7u1gqn7x+umFQ93521vFTcZJIQCWjB1PFbdAZHhIdklMoDxb0Hu3vhJj1wivp+j+nIPG0btQsC8Sk92Cg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"d9e430203fa568b51d3f767255ba236e9ef26fd536bbb77f679b7b5d074f7ab7","last_reissued_at":"2026-07-05T10:48:14.690091Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:48:14.690091Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Plato: Plan to Efficiently Decode for Large Language Model Inference","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Atul Prakash, Danyang Zhuo, Feng Qian, Haizhong Zheng, Matthew Lentz, Qingzhao Zhang, Shuowei Jin, Xueshen Liu, Yongji Wu, Z. Morley Mao","submitted_at":"2024-02-19T16:47:04Z","abstract_excerpt":"Large language models (LLMs) have achieved remarkable success in natural language tasks, but their inference incurs substantial computational and memory overhead. To improve efficiency, parallel decoding methods like Skeleton-of-Thought (SoT) decompose prompts into sub-problems for concurrent processing. However, these methods significantly compromise answer quality by treating semantically linked sub-problems as independent. We propose Plato, a novel approach that co-designs algorithms and systems for semantic-aware parallel decoding. Plato leverages LLMs to organize sub-problems into a depen"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2402.12280","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/2402.12280/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":"2402.12280","created_at":"2026-07-05T10:48:14.690145+00:00"},{"alias_kind":"arxiv_version","alias_value":"2402.12280v2","created_at":"2026-07-05T10:48:14.690145+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2402.12280","created_at":"2026-07-05T10:48:14.690145+00:00"},{"alias_kind":"pith_short_12","alias_value":"3HSDAIB7UVUL","created_at":"2026-07-05T10:48:14.690145+00:00"},{"alias_kind":"pith_short_16","alias_value":"3HSDAIB7UVULKHJ7","created_at":"2026-07-05T10:48:14.690145+00:00"},{"alias_kind":"pith_short_8","alias_value":"3HSDAIB7","created_at":"2026-07-05T10:48:14.690145+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2404.14294","citing_title":"A Survey on Efficient Inference for Large Language Models","ref_index":48,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/3HSDAIB7UVULKHJ7OZZFLORDN2","json":"https://pith.science/pith/3HSDAIB7UVULKHJ7OZZFLORDN2.json","graph_json":"https://pith.science/api/pith-number/3HSDAIB7UVULKHJ7OZZFLORDN2/graph.json","events_json":"https://pith.science/api/pith-number/3HSDAIB7UVULKHJ7OZZFLORDN2/events.json","paper":"https://pith.science/paper/3HSDAIB7"},"agent_actions":{"view_html":"https://pith.science/pith/3HSDAIB7UVULKHJ7OZZFLORDN2","download_json":"https://pith.science/pith/3HSDAIB7UVULKHJ7OZZFLORDN2.json","view_paper":"https://pith.science/paper/3HSDAIB7","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2402.12280&json=true","fetch_graph":"https://pith.science/api/pith-number/3HSDAIB7UVULKHJ7OZZFLORDN2/graph.json","fetch_events":"https://pith.science/api/pith-number/3HSDAIB7UVULKHJ7OZZFLORDN2/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/3HSDAIB7UVULKHJ7OZZFLORDN2/action/timestamp_anchor","attest_storage":"https://pith.science/pith/3HSDAIB7UVULKHJ7OZZFLORDN2/action/storage_attestation","attest_author":"https://pith.science/pith/3HSDAIB7UVULKHJ7OZZFLORDN2/action/author_attestation","sign_citation":"https://pith.science/pith/3HSDAIB7UVULKHJ7OZZFLORDN2/action/citation_signature","submit_replication":"https://pith.science/pith/3HSDAIB7UVULKHJ7OZZFLORDN2/action/replication_record"}},"created_at":"2026-07-05T10:48:14.690145+00:00","updated_at":"2026-07-05T10:48:14.690145+00:00"}