{"id":"W4412366363","doi":"10.1080/10875301.2025.2529435","title":"Evaluating the Impact of a Proactive Chat Feature on Reference Questions, Their Complexity and Subject Matter","year":2025,"lang":"en","type":"article","venue":"Internet Reference Services Quarterly","topic":"Digital Communication and Language","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"McGill University","funders":"","keywords":"Feature (linguistics); Computer science; Subject matter; Subject (documents); World Wide Web; Psychology; Information retrieval; Data science; Linguistics","routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":true,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0003264294,0.0002123336,0.0002203432,0.0001156356,0.00008524433,0.000358474,0.001613881,0.00007639436,0.00003646892],"category_scores_gemma":[0.000009343619,0.000122545,0.0000665925,0.000273718,0.0001138015,0.0004942141,0.0002289213,0.0003414662,0.00003299754],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00005993704,"about_ca_system_score_gemma":0.00007094908,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002753104,"about_ca_topic_score_gemma":0.0008616152,"domain_scores_codex":[0.9986925,0.0003139942,0.0002424045,0.0003591379,0.0001917974,0.000200139],"domain_scores_gemma":[0.9982893,0.0002477801,0.0001939128,0.001046156,0.000174226,0.00004859221],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"observational","study_design_scores_codex":[0.0005956072,0.0007666953,0.0127766,0.0003797801,0.0006532119,0.000005222339,0.07144073,0.0000385116,0.003373641,0.4107823,0.000858391,0.4983293],"study_design_scores_gemma":[0.001050086,0.005216148,0.8612509,0.001540701,0.00004143893,0.00003696887,0.006658904,0.07387685,0.00208447,0.04662557,0.0009267165,0.0006912898],"study_design_candidate":"observational","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9388732,0.0002783266,0.001140908,0.0009638399,0.00003905004,0.0003322121,0.00002971326,0.00008188459,0.05826081],"genre_scores_gemma":[0.9978395,0.000008609113,0.0006370948,0.0005729055,0.00000578978,0.00003437552,0.0000214338,0.00000571277,0.000874556],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.8484743,"threshold_uncertainty_score":0.4997241,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.06027548121335283,"score_gpt":0.3701431148807869,"score_spread":0.3098676336674341,"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."}}