{"id":"W7106798635","doi":"10.48448/6we4-mh95","title":"Self-Guided Function Calling in Large Language Models via Stepwise Experience Recall","year":2025,"lang":"","type":"other","venue":"Open MIND","topic":"","field":"","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Alberta","funders":"","keywords":"Function (biology); Task (project management); Recall; Diversity (politics); Language model; Precision and recall","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":[0.002911132,0.003148733,0.001333004,0.001099917,0.0005501686,0.002663541,0.00504417,0.001641853,0.005542138],"category_scores_gemma":[0.01310652,0.001133642,0.002044524,0.0006568146,0.001013343,0.006058751,0.004443087,0.002497447,0.004967029],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.000828743,"about_ca_system_score_gemma":0.001811866,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.007206007,"about_ca_topic_score_gemma":0.01352577,"domain_scores_codex":[0.9971728,0.001036028,0.0001808969,0.0008839184,0.0004960305,0.0002301765],"domain_scores_gemma":[0.9935167,0.003022152,0.0003456847,0.002379067,0.000541854,0.0001945617],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0008732573,0.0007284966,0.01109673,0.0008779293,0.0003815804,0.0007693507,0.001725715,0.1969004,0.02735286,0.00906573,0.03012173,0.7201063],"study_design_scores_gemma":[0.0001255302,0.000259972,0.000631051,0.00004784059,0.00008803168,0.0002165401,0.0002007935,0.9587874,0.01725677,0.01012238,0.01218315,0.0000805451],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.08243254,0.001088675,0.776409,0.0005947562,0.0001043226,0.0002948653,0.001386076,0.1322157,0.005474044],"genre_scores_gemma":[0.4338234,0.0003815081,0.5445989,0.0007197125,0.00007024482,0.0004703062,0.006864258,0.006430892,0.006640779],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.007206007,"threshold_uncertainty_score":0.01854026,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04384854837543284,"score_gpt":0.3355093863027455,"score_spread":0.2916608379273126,"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."}}