{"id":"W4415704567","doi":"10.1016/j.buildenv.2025.113836","title":"Query-Distill: Cloud–fog collaboration with knowledge distillation and confidence-based querying for smart buildings","year":2025,"lang":"en","type":"article","venue":"Building and Environment","topic":"Context-Aware Activity Recognition Systems","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":false,"ca_institutions":"Concordia University","funders":"Natural Sciences and Engineering Research Council of Canada; Concordia University","keywords":"Robustness (evolution); Occupancy; Software deployment; Efficient energy use; Building automation; Electricity; Distillation; Activity recognition","routes":{"ca_aff":true,"ca_fund":true,"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":[0.0003714375,0.0001656934,0.000201331,0.0001223286,0.0002899727,0.0002637161,0.0001178802,0.00006604161,0.000002584102],"category_scores_gemma":[0.00003861755,0.0001552179,0.00002625026,0.0001733662,0.00008326653,0.0002885951,0.00008588298,0.00006514177,0.000001994673],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001285896,"about_ca_system_score_gemma":0.00006539756,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00007059347,"about_ca_topic_score_gemma":0.0000448844,"domain_scores_codex":[0.9989027,0.00006155433,0.0002158224,0.0004957293,0.0001362094,0.0001879521],"domain_scores_gemma":[0.9992401,0.0002997752,0.0001166375,0.000226941,0.00004251932,0.00007400379],"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":[0.0003676843,0.0003941113,0.05983251,0.0008517929,0.0002439382,0.000009946441,0.003199722,0.002908616,0.03138015,0.1297255,0.002882395,0.7682036],"study_design_scores_gemma":[0.008200783,0.001083104,0.05913673,0.002078264,0.0002625198,0.00005880347,0.0009569422,0.4562302,0.03604013,0.01148893,0.4224484,0.002015215],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.2013577,0.0004619693,0.7962536,0.001021023,0.0001827028,0.0004561454,0.000008654502,0.00007176412,0.0001864031],"genre_scores_gemma":[0.9820649,0.00003895007,0.01740566,0.00009884027,0.00004334184,0.0001268071,0.000007659417,0.000009470392,0.000204385],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.7807072,"threshold_uncertainty_score":0.6329601,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01243880778592586,"score_gpt":0.2478960926958368,"score_spread":0.235457284909911,"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."}}