{"id":"W4406457848","doi":"10.1109/bigdata62323.2024.10825619","title":"From Graph Paths to Natural Language: Enhancing LLM Reasoning for Multi-choice Question-Answering Tasks","year":2024,"lang":"en","type":"article","venue":"","topic":"Topic Modeling","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Carleton University","funders":"","keywords":"Question answering; Computer science; Natural language; Graph; Artificial intelligence; Natural language processing; Theoretical computer science","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":[0.001999849,0.001011406,0.0005192012,0.002163307,0.0005501023,0.001283634,0.001653693,0.001252191,0.004478403],"category_scores_gemma":[0.009327657,0.0004567631,0.001710236,0.001561688,0.0008623399,0.005971943,0.002846968,0.0019021,0.001132052],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001175133,"about_ca_system_score_gemma":0.001768807,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01064615,"about_ca_topic_score_gemma":0.01705656,"domain_scores_codex":[0.9983141,0.0007853823,0.00009711712,0.00046385,0.0002464934,0.00009304635],"domain_scores_gemma":[0.9968264,0.002179157,0.0001674325,0.0004406747,0.0002730524,0.0001132361],"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":[0.0004979974,0.000642437,0.005585887,0.001037643,0.0002148181,0.000670146,0.002532133,0.09348746,0.0205649,0.07360602,0.02293148,0.7782291],"study_design_scores_gemma":[0.00007357198,0.00009188591,0.0008022304,0.00006489838,0.0000935349,0.0001613426,0.0003079343,0.8401946,0.009522557,0.1325727,0.01606881,0.00004600611],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.02087185,0.0004170099,0.9663293,0.001094851,0.0000468508,0.0002487512,0.001037864,0.008177176,0.001776325],"genre_scores_gemma":[0.2457099,0.0003312328,0.7482513,0.0003912406,0.00004991581,0.0002681707,0.003126236,0.0003279385,0.001544062],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.01064615,"threshold_uncertainty_score":0.02116841,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01681412233054059,"score_gpt":0.3105731279335161,"score_spread":0.2937590056029755,"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."}}