{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:3JXEDU3LLM5DROMO5RCETUT7LO","short_pith_number":"pith:3JXEDU3L","schema_version":"1.0","canonical_sha256":"da6e41d36b5b3a38b98eec4449d27f5ba96bc65a56d917365bae6cdb4ee6a71c","source":{"kind":"arxiv","id":"2412.15797","version":1},"attestation_state":"computed","paper":{"title":"Ensembling Large Language Models with Process Reward-Guided Tree Search for Better Complex Reasoning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Edward Choi, Sungjin Park, Xiao Liu, Yeyun Gong","submitted_at":"2024-12-20T11:14:29Z","abstract_excerpt":"Despite recent advances in large language models, open-source models often struggle to consistently perform well on complex reasoning tasks. Existing ensemble methods, whether applied at the token or output levels, fail to address these challenges. In response, we present Language model Ensemble with Monte Carlo Tree Search (LE-MCTS), a novel framework for process-level ensembling of language models. LE-MCTS formulates step-by-step reasoning with an ensemble of language models as a Markov decision process. In this framework, states represent intermediate reasoning paths, while actions consist "},"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":"2412.15797","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-12-20T11:14:29Z","cross_cats_sorted":[],"title_canon_sha256":"81537d0927aed9709b90cb1389e66134aafe69e7c2d3e06d30692fb4e056721c","abstract_canon_sha256":"aab63d94b08772689abedf55e3adc2260115ab3c001056f2db46a2670107f095"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:52:30.151987Z","signature_b64":"1Sex9+z8apslKksiLrMUFiWDtscFiYNTe2GhgBXiECAz20VBXVghgFX2076nrHN5L85QD9n/Jq3GFDzAT8MKAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"da6e41d36b5b3a38b98eec4449d27f5ba96bc65a56d917365bae6cdb4ee6a71c","last_reissued_at":"2026-07-05T09:52:30.151534Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:52:30.151534Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Ensembling Large Language Models with Process Reward-Guided Tree Search for Better Complex Reasoning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Edward Choi, Sungjin Park, Xiao Liu, Yeyun Gong","submitted_at":"2024-12-20T11:14:29Z","abstract_excerpt":"Despite recent advances in large language models, open-source models often struggle to consistently perform well on complex reasoning tasks. Existing ensemble methods, whether applied at the token or output levels, fail to address these challenges. In response, we present Language model Ensemble with Monte Carlo Tree Search (LE-MCTS), a novel framework for process-level ensembling of language models. LE-MCTS formulates step-by-step reasoning with an ensemble of language models as a Markov decision process. In this framework, states represent intermediate reasoning paths, while actions consist "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2412.15797","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/2412.15797/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":"2412.15797","created_at":"2026-07-05T09:52:30.151600+00:00"},{"alias_kind":"arxiv_version","alias_value":"2412.15797v1","created_at":"2026-07-05T09:52:30.151600+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2412.15797","created_at":"2026-07-05T09:52:30.151600+00:00"},{"alias_kind":"pith_short_12","alias_value":"3JXEDU3LLM5D","created_at":"2026-07-05T09:52:30.151600+00:00"},{"alias_kind":"pith_short_16","alias_value":"3JXEDU3LLM5DROMO","created_at":"2026-07-05T09:52:30.151600+00:00"},{"alias_kind":"pith_short_8","alias_value":"3JXEDU3L","created_at":"2026-07-05T09:52:30.151600+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":4,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.11470","citing_title":"The Periodic Table of LLM Reasoning: A Structured Survey of Reasoning Paradigms, Methods, and Failure Modes","ref_index":187,"is_internal_anchor":false},{"citing_arxiv_id":"2502.18036","citing_title":"Harnessing Multiple Large Language Models: A Survey on LLM Ensemble","ref_index":39,"is_internal_anchor":false},{"citing_arxiv_id":"2510.08592","citing_title":"Less Diverse, Less Safe: The Indirect But Pervasive Risk of Test-Time Scaling in Large Language Models","ref_index":14,"is_internal_anchor":false},{"citing_arxiv_id":"2512.23213","citing_title":"Scoring, Reasoning, and Selecting the Best! Ensembling Large Language Models via a Peer-Review Process","ref_index":40,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/3JXEDU3LLM5DROMO5RCETUT7LO","json":"https://pith.science/pith/3JXEDU3LLM5DROMO5RCETUT7LO.json","graph_json":"https://pith.science/api/pith-number/3JXEDU3LLM5DROMO5RCETUT7LO/graph.json","events_json":"https://pith.science/api/pith-number/3JXEDU3LLM5DROMO5RCETUT7LO/events.json","paper":"https://pith.science/paper/3JXEDU3L"},"agent_actions":{"view_html":"https://pith.science/pith/3JXEDU3LLM5DROMO5RCETUT7LO","download_json":"https://pith.science/pith/3JXEDU3LLM5DROMO5RCETUT7LO.json","view_paper":"https://pith.science/paper/3JXEDU3L","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2412.15797&json=true","fetch_graph":"https://pith.science/api/pith-number/3JXEDU3LLM5DROMO5RCETUT7LO/graph.json","fetch_events":"https://pith.science/api/pith-number/3JXEDU3LLM5DROMO5RCETUT7LO/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/3JXEDU3LLM5DROMO5RCETUT7LO/action/timestamp_anchor","attest_storage":"https://pith.science/pith/3JXEDU3LLM5DROMO5RCETUT7LO/action/storage_attestation","attest_author":"https://pith.science/pith/3JXEDU3LLM5DROMO5RCETUT7LO/action/author_attestation","sign_citation":"https://pith.science/pith/3JXEDU3LLM5DROMO5RCETUT7LO/action/citation_signature","submit_replication":"https://pith.science/pith/3JXEDU3LLM5DROMO5RCETUT7LO/action/replication_record"}},"created_at":"2026-07-05T09:52:30.151600+00:00","updated_at":"2026-07-05T09:52:30.151600+00:00"}