{"id":"W4415499254","doi":"10.54195/irrj.22626","title":"Effectiveness of In-Context Learning for Due Diligence","year":2025,"lang":"en","type":"article","venue":"Information Retrieval Research","topic":"Topic Modeling","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Nederlandse Organisatie voor Wetenschappelijk Onderzoek; Universiteit van Amsterdam; Canadian Institute of Steel Construction","keywords":"Due diligence; Dependency (UML); Python (programming language); Diligence; Task (project management); Natural language","routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.01298335,0.001665102,0.001996007,0.002079459,0.00250744,0.002987471,0.003172531,0.002200699,0.004642055],"category_scores_gemma":[0.07293963,0.0006855087,0.001360868,0.001294796,0.001832103,0.008217785,0.003446486,0.004526882,0.001861828],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.003029839,"about_ca_system_score_gemma":0.003849484,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.009666372,"about_ca_topic_score_gemma":0.01062861,"domain_scores_codex":[0.990671,0.004513618,0.000466565,0.002335895,0.001481972,0.0005308167],"domain_scores_gemma":[0.9584615,0.02788567,0.001768379,0.007398293,0.003083107,0.001403052],"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":[0.001211853,0.001005711,0.02672313,0.0005696972,0.0003263836,0.0004436578,0.00182636,0.1910476,0.006994073,0.02621812,0.02137245,0.722261],"study_design_scores_gemma":[0.0000897169,0.0004579606,0.005100803,0.00009507784,0.0001062938,0.0003071153,0.00039276,0.9286689,0.007503203,0.04865748,0.008527126,0.00009357809],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.2412555,0.008216811,0.7157658,0.003783252,0.000658663,0.0005434104,0.0007716794,0.01074978,0.01825514],"genre_scores_gemma":[0.9022862,0.0006109194,0.08897599,0.000603479,0.0002520184,0.0001482627,0.001235589,0.0006077993,0.005279867],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.01298335,"threshold_uncertainty_score":0.06866336,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04890007199663601,"score_gpt":0.383754969905804,"score_spread":0.3348548979091681,"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."}}