{"id":"W4413963077","doi":"10.1111/2041-210x.70140","title":"Optimizing image capture for computer vision‐powered taxonomic identification and trait recognition of biodiversity specimens","year":2025,"lang":"en","type":"article","venue":"Methods in Ecology and Evolution","topic":"Species Distribution and Climate Change","field":"Environmental Science","cited_by":5,"is_retracted":false,"has_abstract":true,"ca_institutions":"McGill University; University of Guelph; Vector Institute","funders":"Office of Advanced Cyberinfrastructure; Natural Sciences and Engineering Research Council of Canada; National Institute of Food and Agriculture; Arkansas NSF EPSCoR; Battelle; Office of International Science and Engineering; U.S. Department of Agriculture; Office of Experimental Program to Stimulate Competitive Research; National Science Foundation","keywords":"Biodiversity; Trait; Identification (biology); Artificial intelligence; Taxonomic rank; Biology; Pattern recognition (psychology); Computer science; Computer vision; Ecology; Taxon","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":[0.0005822958,0.00005687164,0.000113027,0.00006135005,0.0001018822,0.000009356737,0.00004041676,0.00009938539,0.0005800613],"category_scores_gemma":[0.00005371038,0.00006076408,0.00002288998,0.0001020997,0.0001971025,0.0001160625,0.00006629846,0.00005209526,0.00000815844],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0002353787,"about_ca_system_score_gemma":0.000004040323,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00006183352,"about_ca_topic_score_gemma":0.0002903396,"domain_scores_codex":[0.9994051,0.0001209968,0.0001561035,0.0001917794,0.00002412485,0.0001018205],"domain_scores_gemma":[0.9997205,0.0001166637,0.00007342979,0.00005806411,0.00001290376,0.00001846694],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"observational","study_design_scores_codex":[0.001042382,0.000816723,0.4294018,0.0003395538,0.00009590959,0.000003578376,0.002167008,0.0002990353,0.3995968,0.004356528,0.02496513,0.1369156],"study_design_scores_gemma":[0.000569522,0.00004590221,0.9916716,0.000007253467,0.00002174788,0.000001887623,0.000622767,0.002627899,0.00167792,0.001782261,0.000907774,0.00006348461],"study_design_candidate":"observational","study_design_consensus":"observational","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.7825203,0.00006372243,0.21541,0.0004242742,0.0002436098,0.000275311,0.00009611261,0.00001075149,0.0009559519],"genre_scores_gemma":[0.7508289,0.0001085034,0.2486854,0.0001323449,0.000009951765,0.00002904149,0.0001076258,0.000002367761,0.000095873],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.5622698,"threshold_uncertainty_score":0.6351264,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03222701735270504,"score_gpt":0.3141453723850346,"score_spread":0.2819183550323296,"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."}}