{"id":"W4404399990","doi":"10.48550/arxiv.2411.08010","title":"ExpressivityBench: Can LLMs Communicate Implicitly?","year":2024,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Biomedical Text Mining and Ontologies","field":"Biochemistry, Genetics and Molecular Biology","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Atomic Energy of Canada Limited","keywords":"Computer science; Business; Computer security","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":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.0001729955,0.0003169075,0.0002940486,0.00009652932,0.0001061327,0.00005043429,0.001037098,0.0007109462,0.00003301887],"category_scores_gemma":[0.00006012459,0.000322358,0.0002566379,0.0001555582,0.0003651511,0.000001328067,0.003949346,0.0007446275,0.00005637861],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00004780668,"about_ca_system_score_gemma":0.0001897124,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0002838337,"about_ca_topic_score_gemma":0.0002331266,"domain_scores_codex":[0.9984159,0.0001280183,0.000170552,0.0008728863,0.00005852135,0.0003541458],"domain_scores_gemma":[0.9983022,0.00003093001,0.0001188753,0.001318239,0.00006324463,0.0001665126],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"bench_or_experimental","study_design_gemma":"not_applicable","study_design_scores_codex":[0.001837962,0.002510353,0.02878496,0.004966274,0.008538977,0.005490668,0.003469222,0.05486235,0.3474351,0.09504052,0.3424196,0.104644],"study_design_scores_gemma":[0.005031383,0.001967744,0.00691483,0.002377711,0.002055676,0.0001767615,0.003405302,0.03689362,0.0865246,0.2505829,0.5959653,0.008104176],"study_design_candidate":"not_applicable","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9870626,0.0019345,0.002502697,0.0004560956,0.0006328147,0.0001901349,0.0001821826,0.0001556489,0.006883323],"genre_scores_gemma":[0.9934109,0.0009788258,0.0002777998,0.000184469,0.0001815228,0.000002494461,0.0002187871,0.00003526448,0.00470992],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.2609105,"threshold_uncertainty_score":0.9999229,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.08480621804238049,"score_gpt":0.2175119563183353,"score_spread":0.1327057382759548,"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."}}