{"id":"W4403180322","doi":"10.2139/ssrn.4978367","title":"Fishllm: Multimodal Instruction Tuning with Large Language Models for Fish Classification and Detection","year":2024,"lang":"en","type":"preprint","venue":"SSRN Electronic Journal","topic":"Water Quality Monitoring Technologies","field":"Environmental Science","cited_by":0,"is_retracted":false,"has_abstract":false,"ca_institutions":"University of Alberta","funders":"","keywords":"Fish <Actinopterygii>; Computer science; Artificial intelligence; Natural language processing; Linguistics; Fishery; Biology; Philosophy","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":["research_integrity"],"consensus_categories":[],"category_scores_codex":[0.001125817,0.0002375749,0.0001906747,0.0001258987,0.0002489903,0.0001784038,0.0002676552,0.0003029018,0.000004612882],"category_scores_gemma":[0.00004150161,0.0002036886,0.00007055209,0.0001158301,0.00009466412,0.0002322157,0.0004881567,0.002826794,0.000005822094],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001982455,"about_ca_system_score_gemma":0.0001591215,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0003232194,"about_ca_topic_score_gemma":0.002378616,"domain_scores_codex":[0.9978356,0.00005678627,0.0002578283,0.0004761266,0.0002736919,0.001099926],"domain_scores_gemma":[0.9994383,0.0000271479,0.0002087739,0.0002585344,0.00001960482,0.00004760969],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"theoretical_or_conceptual","study_design_scores_codex":[0.001043299,0.0003658013,0.01342155,0.000762107,0.001305056,0.00002814924,0.01100032,0.03299462,0.07893961,0.0473856,0.0003944459,0.8123595],"study_design_scores_gemma":[0.001234165,0.0005269533,0.003318257,0.0002260089,0.0002111783,0.0005038047,0.008508467,0.1254266,0.00759771,0.851347,0.0003968143,0.0007030104],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9103369,0.0002617413,0.08761903,0.0007223673,0.0003493254,0.0003630688,0.00002161151,0.0002403735,0.00008564867],"genre_scores_gemma":[0.996288,0.000514991,0.002541895,0.00001270904,0.0002241599,0.00008009346,0.00001855267,0.00004440246,0.000275202],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.8116564,"threshold_uncertainty_score":0.9994738,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01885318915807857,"score_gpt":0.2585033517656453,"score_spread":0.2396501626075667,"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."}}