{"id":"W4386728930","doi":"10.1145/3604915.3608845","title":"Large Language Models are Competitive Near Cold-start Recommenders for Language- and Item-based Preferences","year":2023,"lang":"en","type":"article","venue":"","topic":"Recommender Systems and Techniques","field":"Computer Science","cited_by":112,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto","funders":"","keywords":"Leverage (statistics); Computer science; Preference; Cold start (automotive); Natural language processing; Artificial intelligence; Dialog box; Task (project management); Recommender system; Variety (cybernetics); Preference learning; Language model; Machine learning; World Wide Web","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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.000580503,0.0001786033,0.0002615884,0.0001317418,0.000204313,0.0002820472,0.0004345982,0.00008410967,0.00002667596],"category_scores_gemma":[0.00002991473,0.0001486117,0.00007259838,0.0003041459,0.00003032059,0.0003760647,0.00018944,0.00009615685,0.000009804647],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00002820115,"about_ca_system_score_gemma":0.00005160687,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0001328278,"about_ca_topic_score_gemma":0.0003857818,"domain_scores_codex":[0.9986196,0.00009473815,0.0002247,0.0004362317,0.0002070719,0.0004176184],"domain_scores_gemma":[0.9989586,0.0003664047,0.0001040577,0.000390555,0.00006928331,0.0001111388],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00006433018,0.000459912,0.006505008,0.0008855216,0.0002275461,0.0001258421,0.05752852,0.0001889959,0.001615351,0.5980273,0.3064411,0.02793059],"study_design_scores_gemma":[0.001684392,0.0003542851,0.0009086608,0.0002473701,0.00001577427,0.00000609583,0.03077735,0.9149851,0.009420553,0.008620623,0.03224055,0.0007391974],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.06761598,0.000239343,0.909421,0.005286549,0.0003241779,0.001259466,0.0001559551,0.002159762,0.01353782],"genre_scores_gemma":[0.9633421,0.00002319018,0.03400058,0.001194343,0.00003912621,0.0001999828,0.00003147582,0.00001882433,0.00115032],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9147962,"threshold_uncertainty_score":0.6060209,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04197899489953127,"score_gpt":0.2851031101057967,"score_spread":0.2431241152062654,"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."}}