{"id":"W4396778270","doi":"10.3389/frai.2024.1255566","title":"Leveraging diffusion models for unsupervised out-of-distribution detection on image manifold","year":2024,"lang":"en","type":"article","venue":"Frontiers in Artificial Intelligence","topic":"Domain Adaptation and Few-Shot Learning","field":"Computer Science","cited_by":2,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Division of Materials Research; Materials Research Science and Engineering Center, Harvard University; Natural Sciences and Engineering Research Council of Canada; Cornell Center for Materials Research; Defense Advanced Research Projects Agency; National Science Foundation","keywords":"Artificial intelligence; Computer science; Pattern recognition (psychology); Domain (mathematical analysis); Inference; Subspace topology; Image (mathematics); Manifold (fluid mechanics); Computer vision; Mathematics","routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0005311594,0.0001434071,0.0001681546,0.0002587783,0.0001198281,0.0002088143,0.0003609169,0.00007918256,0.000007159928],"category_scores_gemma":[0.00009981277,0.0001480033,0.000102334,0.0005020053,0.00004240082,0.0005271464,0.00005824862,0.0001985782,0.00002612046],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001537394,"about_ca_system_score_gemma":0.00004433467,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00001961577,"about_ca_topic_score_gemma":0.00001108184,"domain_scores_codex":[0.9985498,0.00006866505,0.0004178121,0.0004493479,0.0002473656,0.0002670285],"domain_scores_gemma":[0.9994293,0.0001237774,0.00006590405,0.0002455382,0.00008087605,0.00005464496],"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":[0.00008215776,0.00008402715,0.00001459841,0.00007117706,0.00001250412,0.00001057565,0.003843859,0.0217892,0.009857843,0.1395125,0.0003548598,0.8243667],"study_design_scores_gemma":[0.00003167087,0.0000845514,0.00002176168,0.00009898085,0.000004181538,7.504506e-7,0.0005713966,0.8406475,0.04977934,0.1075012,0.001125341,0.0001333312],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.004572966,0.0001519377,0.9905335,0.0002684665,0.003570192,0.0003047124,0.000007706336,0.0001636396,0.000426928],"genre_scores_gemma":[0.9438202,0.0000396625,0.05585272,0.00005514126,0.00007784218,0.00003397513,0.00001383658,0.00001350255,0.00009311497],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9392473,"threshold_uncertainty_score":0.6035399,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.05809887125542123,"score_gpt":0.28941310745886,"score_spread":0.2313142362034387,"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."}}