{"id":"W7077852133","doi":"10.48448/5k8n-qv56","title":"Ensembling Large Language Models with Process Reward-Guided Tree Search for Better Complex Reasoning","year":2025,"lang":"en","type":"other","venue":"Open MIND","topic":"Geochemistry and Geologic Mapping","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Kootenay Association for Science & Technology","funders":"","keywords":"Language model; Model-based reasoning; Reasoning system; Tree (set theory); Opportunistic reasoning; Abductive reasoning; Qualitative reasoning; Case-based reasoning; Ensemble learning; Process (computing)","routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.002359312,0.001625454,0.001606507,0.001652784,0.0008120934,0.001729839,0.002186644,0.001585635,0.002815677],"category_scores_gemma":[0.01112921,0.0006722459,0.001993391,0.001362558,0.0009089551,0.003516929,0.001879566,0.003091631,0.001347585],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001534244,"about_ca_system_score_gemma":0.002692065,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01122368,"about_ca_topic_score_gemma":0.02058241,"domain_scores_codex":[0.998497,0.0005592491,0.00009275929,0.0004355267,0.0002887494,0.0001266697],"domain_scores_gemma":[0.9956938,0.003152395,0.000187703,0.0004049563,0.0004026975,0.0001583826],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0002508312,0.0002446111,0.004073741,0.0002386714,0.0002373735,0.0003340765,0.0003131385,0.7467253,0.00402705,0.01977979,0.01055129,0.2132242],"study_design_scores_gemma":[0.000009415921,0.00001117133,0.00005816567,0.000006345875,0.000009830371,0.00001614066,0.000009411746,0.9894599,0.0004691342,0.009535722,0.0004097873,0.000005009106],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.04664455,0.001336847,0.9397341,0.0008924283,0.0001395516,0.00009044644,0.0005761374,0.008453739,0.002132161],"genre_scores_gemma":[0.5739852,0.0005789271,0.4163602,0.000859708,0.000160649,0.0002297162,0.003821385,0.0009697053,0.003034501],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.01122368,"threshold_uncertainty_score":0.02231669,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.06674403635020607,"score_gpt":0.3372735317888759,"score_spread":0.2705294954386699,"validation_status":"score_only:v0-immature-baseline","note":"Baseline scores from an immature model (maturity gate not passed). Scores rank; they never assert a category."}}