{"id":"W4392669730","doi":"10.18653/v1/2023.eval4nlp-1.16","title":"Reference-Free Summarization Evaluation with Large Language Models","year":2023,"lang":"en","type":"article","venue":"","topic":"Topic Modeling","field":"Computer Science","cited_by":5,"is_retracted":false,"has_abstract":true,"ca_institutions":"Carleton University; National Research Council Canada","funders":"Natural Sciences and Engineering Research Council of Canada","keywords":"Automatic summarization; Computer science; Natural language processing; Multi-document summarization; Artificial intelligence","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.01149007,0.002696443,0.001601018,0.002266319,0.0009284352,0.002192643,0.002002521,0.002216373,0.004413959],"category_scores_gemma":[0.039738,0.0005028707,0.001126296,0.001570631,0.000720416,0.003973198,0.002500739,0.001892561,0.003164494],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001203178,"about_ca_system_score_gemma":0.001353248,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.003044166,"about_ca_topic_score_gemma":0.003886316,"domain_scores_codex":[0.9851664,0.009841145,0.0009460261,0.002036097,0.001674903,0.0003354595],"domain_scores_gemma":[0.970303,0.01808246,0.001049222,0.003798056,0.00602075,0.0007465422],"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":[0.003843812,0.001283542,0.004939765,0.002728678,0.0008979411,0.0008152232,0.002598559,0.22611,0.07528121,0.007120096,0.04395932,0.6304219],"study_design_scores_gemma":[0.0004242603,0.001914385,0.002898298,0.00009235091,0.0002431316,0.0002480858,0.0006242316,0.9224698,0.05244535,0.006447557,0.01206098,0.0001316637],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.2286416,0.003270511,0.6881353,0.0006524301,0.0006873858,0.0009909949,0.004416373,0.06612419,0.007081217],"genre_scores_gemma":[0.6700061,0.000438462,0.302716,0.0002795084,0.0002084702,0.0007775879,0.01857862,0.002712889,0.004282323],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.01149007,"threshold_uncertainty_score":0.06076604,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.06768538163856708,"score_gpt":0.2967039691196833,"score_spread":0.2290185874811162,"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."}}