{"id":"W4413391566","doi":"10.18260/1-2--55492","title":"Automated Grading of Engineering Mechanics Assignments Using Large Language Models and Computer Vision: A Work in Progress","year":2025,"lang":"en","type":"article","venue":"","topic":"Natural Language Processing Techniques","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":false,"ca_institutions":"University of Alberta","funders":"","keywords":"Computer science; Grading (engineering); Work (physics); Artificial intelligence; Natural language processing; Engineering drawing; Mechanical engineering; Engineering; Civil engineering","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":[0.002528913,0.001703601,0.001899647,0.003529464,0.0009565281,0.004249564,0.00379216,0.001469997,0.007540472],"category_scores_gemma":[0.01010184,0.0006494722,0.001950804,0.001803068,0.0007292326,0.004647478,0.001753468,0.002543066,0.005486092],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001685841,"about_ca_system_score_gemma":0.003208659,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.007669572,"about_ca_topic_score_gemma":0.0152788,"domain_scores_codex":[0.9951184,0.001835122,0.000298054,0.0009995209,0.001437825,0.00031095],"domain_scores_gemma":[0.989021,0.003428006,0.0009503622,0.002261274,0.003839457,0.0004998946],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.0002921901,0.001070557,0.003141188,0.0004370284,0.0001271546,0.0001151125,0.0001576428,0.03744296,0.02721384,0.006057071,0.02429534,0.8996499],"study_design_scores_gemma":[0.00006886093,0.0001596567,0.002225497,0.00005766444,0.00005230826,0.0001061301,0.0002501585,0.9425361,0.02042598,0.02226694,0.0117836,0.00006712454],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.04911559,0.0006855421,0.9210765,0.0007412824,0.0002814663,0.0003732323,0.00191084,0.02277811,0.003037515],"genre_scores_gemma":[0.3426468,0.0004699588,0.6374211,0.0003268322,0.0001914141,0.0002684836,0.00917558,0.001306659,0.0081932],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.007669572,"threshold_uncertainty_score":0.02522534,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01006253681591404,"score_gpt":0.2887079602246625,"score_spread":0.2786454234087485,"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."}}