{"id":"W4405010024","doi":"10.1016/j.foodcont.2024.111071","title":"Pixel-level spectral aflatoxin B1 content intelligent prediction via fine-tuning large language model (LLM)","year":2024,"lang":"en","type":"article","venue":"Food Control","topic":"Spectroscopy and Chemometric Analyses","field":"Chemistry","cited_by":11,"is_retracted":false,"has_abstract":false,"ca_institutions":"Ministry of Agriculture","funders":"National Key Research and Development Program of China; National Natural Science Foundation of China","keywords":"Pixel; Aflatoxin; Content (measure theory); Computer science; Fine-tuning; Artificial intelligence; Biological system; Chemistry; Food science; Mathematics; Physics; Biology","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0004741536,0.0007719764,0.0007143895,0.0004195669,0.0003415633,0.0007267753,0.0009869161,0.0007656931,0.001987227],"category_scores_gemma":[0.001244456,0.0003222493,0.0007896944,0.0003840507,0.0002119504,0.0008132507,0.0006700142,0.001212926,0.001081427],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0005598,"about_ca_system_score_gemma":0.0006862761,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.009915921,"about_ca_topic_score_gemma":0.01523179,"domain_scores_codex":[0.9997702,0.00003664811,0.00001254097,0.0001061843,0.00003492711,0.00003945701],"domain_scores_gemma":[0.9996618,0.0001671701,0.00002726727,0.00003759382,0.00008679348,0.00001940381],"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":[0.0006564754,0.0005424321,0.005147311,0.0001580327,0.0002322735,0.0002068105,0.00009610585,0.4060631,0.05122731,0.001746517,0.008380064,0.5255436],"study_design_scores_gemma":[0.000004952349,0.00001711786,0.0002402429,0.000001837502,0.000009881531,0.000008269748,0.000004543152,0.9968181,0.002271201,0.0004162228,0.0002037031,0.00000404165],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1835127,0.001318025,0.7921764,0.0008771088,0.0002484796,0.0000789166,0.001180911,0.01764843,0.002959085],"genre_scores_gemma":[0.8678638,0.0001999158,0.1258591,0.0003889534,0.00006038411,0.0001068615,0.001550393,0.0002628814,0.003707729],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.009915921,"threshold_uncertainty_score":0.01971638,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0383959038331966,"score_gpt":0.2694659419521947,"score_spread":0.2310700381189981,"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."}}