{"id":"W4403244299","doi":"10.1039/d4sc04630g","title":"Automated electrosynthesis reaction mining with multimodal large language models (MLLMs)","year":2024,"lang":"en","type":"article","venue":"Chemical Science","topic":"Machine Learning in Materials Science","field":"Materials Science","cited_by":18,"is_retracted":false,"has_abstract":true,"ca_institutions":"Vector Institute; University of Toronto","funders":"Office of Science; Canada First Research Excellence Fund; University of Toronto; U.S. Department of Energy; University of Minnesota; Nanyang Technological University; Ministry of Education - Singapore; Canadian Institute for Advanced Research","keywords":"Electrosynthesis; Computer science; Natural language processing; Data science; Chemistry; Electrochemistry","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":[0.001858955,0.0002225106,0.0002145648,0.000157092,0.0003202492,0.0007026802,0.00089274,0.0000793471,0.0003611725],"category_scores_gemma":[0.0004193808,0.0001605873,0.00004039812,0.001220348,0.0006734642,0.001273453,0.0002166271,0.0001947614,0.0003268165],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0002198981,"about_ca_system_score_gemma":0.0002638243,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0001029828,"about_ca_topic_score_gemma":0.000005634663,"domain_scores_codex":[0.9968739,0.0000577539,0.0002602314,0.0009709528,0.0009509603,0.000886272],"domain_scores_gemma":[0.9990187,0.0001835908,0.00007816505,0.0004146336,0.00008601946,0.00021895],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"bench_or_experimental","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.00001983815,0.00002858575,0.00003071186,0.00003216903,0.000001531399,0.00002663745,0.0009902054,0.0005578983,0.9970678,0.0006076226,0.0001293049,0.0005076698],"study_design_scores_gemma":[0.00007197047,0.00002592308,0.00009406402,0.00007220644,0.000005611872,0.00004726052,0.0001015784,0.3987366,0.6004706,0.00009899552,0.0001077056,0.0001674681],"study_design_candidate":"bench_or_experimental","study_design_consensus":"bench_or_experimental","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9897751,0.00007681792,0.004721787,0.0002369828,0.0003740746,0.000141143,0.00001484173,0.001909772,0.002749515],"genre_scores_gemma":[0.9796941,0.000002493124,0.01989674,0.00009478062,0.0001161417,0.00002664751,0.00000435444,0.00002460907,0.0001401815],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.3981787,"threshold_uncertainty_score":0.6775962,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.007185492762374514,"score_gpt":0.2715545567418196,"score_spread":0.264369063979445,"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."}}