{"id":"W4403887899","doi":"10.1007/978-3-031-72949-2_7","title":"Adapt Without Forgetting: Distill Proximity from Dual Teachers in Vision-Language Models","year":2024,"lang":"en","type":"book-chapter","venue":"Lecture notes in computer science","topic":"Domain Adaptation and Few-Shot Learning","field":"Computer Science","cited_by":2,"is_retracted":false,"has_abstract":false,"ca_institutions":"Huawei Technologies (Canada)","funders":"","keywords":"Computer science; Forgetting; Dual (grammatical number); Artificial intelligence; Human–computer interaction; Computer vision; Natural language processing; Computer graphics (images); Cognitive science; Cognitive psychology; Linguistics; Psychology","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":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.001203394,0.0006333693,0.0006238749,0.0009243513,0.0001915971,0.001022776,0.002370133,0.0003765764,0.0000264993],"category_scores_gemma":[0.0001389472,0.0005719981,0.0001588736,0.0007638502,0.0005042758,0.001304074,0.001537046,0.001729777,0.00007390349],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0004387854,"about_ca_system_score_gemma":0.0004826535,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0001883557,"about_ca_topic_score_gemma":0.000357199,"domain_scores_codex":[0.9950622,0.00006874985,0.0006951873,0.002130397,0.001271643,0.0007718136],"domain_scores_gemma":[0.997641,0.0005104892,0.000272118,0.001228055,0.0001078336,0.0002404659],"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":[0.00001567664,0.00006404123,0.000097333,0.00004759238,0.00002350528,0.0005425901,0.0277783,0.1976036,0.0002482624,0.04897228,0.00004469091,0.7245621],"study_design_scores_gemma":[0.0002720354,0.0000732601,0.0001077947,0.0005638978,0.000007006152,0.00002102328,0.000005392866,0.8072524,0.000101839,0.1899201,0.00110196,0.0005732876],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.0007413912,0.0009908082,0.9861073,0.00105397,0.001448981,0.0004798181,0.00001210873,0.0003334763,0.008832163],"genre_scores_gemma":[0.6601434,0.00001722961,0.336752,0.0009686726,0.0003711688,0.00001571632,0.00001710907,0.00006183571,0.001652816],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.7239888,"threshold_uncertainty_score":0.9996731,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0192529939868261,"score_gpt":0.2673125237435382,"score_spread":0.2480595297567121,"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."}}