{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:3JXEDU3LLM5DROMO5RCETUT7LO","short_pith_number":"pith:3JXEDU3L","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"},"canonical_sha256":"da6e41d36b5b3a38b98eec4449d27f5ba96bc65a56d917365bae6cdb4ee6a71c","source":{"kind":"arxiv","id":"2412.15797","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2412.15797","created_at":"2026-07-05T09:52:30Z"},{"alias_kind":"arxiv_version","alias_value":"2412.15797v1","created_at":"2026-07-05T09:52:30Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2412.15797","created_at":"2026-07-05T09:52:30Z"},{"alias_kind":"pith_short_12","alias_value":"3JXEDU3LLM5D","created_at":"2026-07-05T09:52:30Z"},{"alias_kind":"pith_short_16","alias_value":"3JXEDU3LLM5DROMO","created_at":"2026-07-05T09:52:30Z"},{"alias_kind":"pith_short_8","alias_value":"3JXEDU3L","created_at":"2026-07-05T09:52:30Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:3JXEDU3LLM5DROMO5RCETUT7LO","target":"record","payload":{"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"},"canonical_sha256":"da6e41d36b5b3a38b98eec4449d27f5ba96bc65a56d917365bae6cdb4ee6a71c","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"},"source_kind":"arxiv","source_id":"2412.15797","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T09:52:30Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"2CcMYSzwNzVspwMgaTbAiLf1TJv/X7O7/hbKBmyTVTSt/uy77u9PNxNKT7kjtCYdN4kHQcGdrfX8A+1BYLiRBw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-06T10:31:01.692872Z"},"content_sha256":"a6469754b2cce4551f93e78a1447364c6a6d123faf4299e13505d631e9dc66d8","schema_version":"1.0","event_id":"sha256:a6469754b2cce4551f93e78a1447364c6a6d123faf4299e13505d631e9dc66d8"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:3JXEDU3LLM5DROMO5RCETUT7LO","target":"graph","payload":{"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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T09:52:30Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"xfbvLc3MAli0O7SuP8lvk716YTHC3Q49/n9DDfGSMC4SAp3j4QCgwnDVNKzBA9U/zmjDwQ6HzudLT44z8xENDQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-06T10:31:01.693624Z"},"content_sha256":"6509d54c77f964f1167be0a346733e065003ebfdabb5cc753f58c4a767b758b9","schema_version":"1.0","event_id":"sha256:6509d54c77f964f1167be0a346733e065003ebfdabb5cc753f58c4a767b758b9"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/3JXEDU3LLM5DROMO5RCETUT7LO/bundle.json","state_url":"https://pith.science/pith/3JXEDU3LLM5DROMO5RCETUT7LO/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/3JXEDU3LLM5DROMO5RCETUT7LO/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-08-06T10:31:01Z","links":{"resolver":"https://pith.science/pith/3JXEDU3LLM5DROMO5RCETUT7LO","bundle":"https://pith.science/pith/3JXEDU3LLM5DROMO5RCETUT7LO/bundle.json","state":"https://pith.science/pith/3JXEDU3LLM5DROMO5RCETUT7LO/state.json","well_known_bundle":"https://pith.science/.well-known/pith/3JXEDU3LLM5DROMO5RCETUT7LO/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:3JXEDU3LLM5DROMO5RCETUT7LO","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"aab63d94b08772689abedf55e3adc2260115ab3c001056f2db46a2670107f095","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-12-20T11:14:29Z","title_canon_sha256":"81537d0927aed9709b90cb1389e66134aafe69e7c2d3e06d30692fb4e056721c"},"schema_version":"1.0","source":{"id":"2412.15797","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2412.15797","created_at":"2026-07-05T09:52:30Z"},{"alias_kind":"arxiv_version","alias_value":"2412.15797v1","created_at":"2026-07-05T09:52:30Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2412.15797","created_at":"2026-07-05T09:52:30Z"},{"alias_kind":"pith_short_12","alias_value":"3JXEDU3LLM5D","created_at":"2026-07-05T09:52:30Z"},{"alias_kind":"pith_short_16","alias_value":"3JXEDU3LLM5DROMO","created_at":"2026-07-05T09:52:30Z"},{"alias_kind":"pith_short_8","alias_value":"3JXEDU3L","created_at":"2026-07-05T09:52:30Z"}],"graph_snapshots":[{"event_id":"sha256:6509d54c77f964f1167be0a346733e065003ebfdabb5cc753f58c4a767b758b9","target":"graph","created_at":"2026-07-05T09:52:30Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2412.15797/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"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 ","authors_text":"Edward Choi, Sungjin Park, Xiao Liu, Yeyun Gong","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-12-20T11:14:29Z","title":"Ensembling Large Language Models with Process Reward-Guided Tree Search for Better Complex Reasoning"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2412.15797","kind":"arxiv","version":1},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:a6469754b2cce4551f93e78a1447364c6a6d123faf4299e13505d631e9dc66d8","target":"record","created_at":"2026-07-05T09:52:30Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"aab63d94b08772689abedf55e3adc2260115ab3c001056f2db46a2670107f095","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-12-20T11:14:29Z","title_canon_sha256":"81537d0927aed9709b90cb1389e66134aafe69e7c2d3e06d30692fb4e056721c"},"schema_version":"1.0","source":{"id":"2412.15797","kind":"arxiv","version":1}},"canonical_sha256":"da6e41d36b5b3a38b98eec4449d27f5ba96bc65a56d917365bae6cdb4ee6a71c","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"da6e41d36b5b3a38b98eec4449d27f5ba96bc65a56d917365bae6cdb4ee6a71c","first_computed_at":"2026-07-05T09:52:30.151534Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T09:52:30.151534Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"1Sex9+z8apslKksiLrMUFiWDtscFiYNTe2GhgBXiECAz20VBXVghgFX2076nrHN5L85QD9n/Jq3GFDzAT8MKAg==","signature_status":"signed_v1","signed_at":"2026-07-05T09:52:30.151987Z","signed_message":"canonical_sha256_bytes"},"source_id":"2412.15797","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:a6469754b2cce4551f93e78a1447364c6a6d123faf4299e13505d631e9dc66d8","sha256:6509d54c77f964f1167be0a346733e065003ebfdabb5cc753f58c4a767b758b9"],"state_sha256":"4c418a5f97f243ed30cbba07c1a7cc9e342d7b7b3fdfffc8604e24733c2f695e"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"ryXFWMSkO1YI92Bq7d9rJHXXo9TxZ/l9uST+v/rZzaYWJ0owpzvRvVNWgIp8tSBvRXheSLtsxutOnURGHM2VCA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-06T10:31:01.699527Z","bundle_sha256":"8efc7391699226fe7526a161024864c13de388ab783f1d731c012071da8b4506"}}