{"id":"W6929151390","doi":"10.48448/v2q6-0106","title":"UniSumEval: Towards Unified, Fine-grained, Multi-dimensional Summarization Evaluation for LLMs","year":2024,"lang":"en","type":"other","venue":"Underline Science Inc.","topic":"","field":"","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Kootenay Association for Science & Technology","funders":"","keywords":"Automatic summarization; Benchmark (surveying); Annotation; Context (archaeology); Focus (optics); Quality (philosophy)","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":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow","insufficient_payload"],"consensus_categories":["insufficient_payload"],"category_scores_codex":[0.005559417,0.0007429778,0.000563266,0.002743038,0.0003300895,0.0004335853,0.001138776,0.0005124207,0.001794188],"category_scores_gemma":[0.001798081,0.0006717824,0.000191843,0.003235945,0.001343727,0.0003929669,0.000391473,0.0004385335,0.005364274],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001071168,"about_ca_system_score_gemma":0.003970577,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0005354206,"about_ca_topic_score_gemma":0.002540095,"domain_scores_codex":[0.992977,0.0001708132,0.0007191734,0.001892383,0.003280251,0.0009604154],"domain_scores_gemma":[0.9964857,0.0001147088,0.0005571141,0.001036908,0.001485971,0.0003195851],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"not_applicable","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0000790677,0.0004293941,0.00002760865,0.0003848652,0.0001739583,0.00001039866,0.0002171955,0.00435963,0.01295857,0.01809506,0.9330475,0.03021677],"study_design_scores_gemma":[0.001864502,0.0001602898,0.00003503444,0.0005041591,0.0004417897,0.00001379924,0.00007695646,0.6782177,0.0006419979,0.004556112,0.3125801,0.0009075512],"study_design_candidate":"not_applicable","study_design_consensus":null,"genre_codex":"other","genre_gemma":"other","genre_scores_codex":[0.0009236502,0.01219497,0.1654791,0.006046143,0.03165276,0.03365791,0.01628192,0.01235076,0.7214128],"genre_scores_gemma":[0.01333607,0.00003179526,0.1238989,0.0004096205,0.001794456,0.0006266095,0.007681535,0.003323365,0.8488976],"genre_candidate":"other","genre_consensus":"other","teacher_disagreement_score":0.673858,"threshold_uncertainty_score":0.9995733,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.08525053730682784,"score_gpt":0.3858531518827633,"score_spread":0.3006026145759355,"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."}}