{"id":"W4404781665","doi":"10.18653/v1/2024.findings-emnlp.192","title":"HoneyComb: A Flexible LLM-Based Agent System for Materials Science","year":2024,"lang":"en","type":"article","venue":"","topic":"Mineral Processing and Grinding","field":"Engineering","cited_by":27,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Natural Sciences and Engineering Research Council of Canada; Canadian Institute for Advanced Research","keywords":"Honeycomb; Computer science; Materials science; Composite material","routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001601141,0.001076622,0.0006917083,0.0007644133,0.0008375196,0.001913403,0.003781588,0.001763555,0.009172487],"category_scores_gemma":[0.005502224,0.0007924761,0.001242888,0.0005209983,0.0009899111,0.003771635,0.004886298,0.002338957,0.003483702],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0009961672,"about_ca_system_score_gemma":0.00187394,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.005881916,"about_ca_topic_score_gemma":0.01083693,"domain_scores_codex":[0.9993419,0.0001846689,0.00007982156,0.0001703261,0.0001779527,0.00004524716],"domain_scores_gemma":[0.9986021,0.0006219218,0.00009985857,0.0004218766,0.0001500698,0.0001040896],"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":[0.001393374,0.0008215797,0.005830914,0.002502566,0.0005864084,0.001454879,0.001488922,0.2587007,0.03793301,0.1204881,0.1338015,0.434998],"study_design_scores_gemma":[0.00013549,0.00008436691,0.0002926012,0.00007123524,0.00006166214,0.0001604846,0.00007835877,0.8768657,0.007815515,0.03992581,0.07444848,0.00006029535],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.007397092,0.0003899265,0.9107326,0.0007235461,0.0001880451,0.000478185,0.001652475,0.07272346,0.00571462],"genre_scores_gemma":[0.1126906,0.0003693525,0.8723212,0.0006172014,0.00005353502,0.000999424,0.004019712,0.002647263,0.006281793],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.009172487,"threshold_uncertainty_score":0.03068507,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0217652522255466,"score_gpt":0.2610658024384261,"score_spread":0.2393005502128795,"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."}}