{"id":"W3140853032","doi":"10.1007/978-3-030-72240-1_11","title":"Comparing Score Aggregation Approaches for Document Retrieval with Pretrained Transformers","year":2021,"lang":"en","type":"book-chapter","venue":"Lecture notes in computer science","topic":"Topic Modeling","field":"Computer Science","cited_by":16,"is_retracted":false,"has_abstract":false,"ca_institutions":"University of Waterloo","funders":"","keywords":"Computer science; Transformer; Artificial intelligence; Question answering; Information retrieval; Machine learning","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"],"consensus_categories":[],"category_scores_codex":[0.0008778077,0.0004998435,0.0005789372,0.0004618907,0.0002719705,0.0006823228,0.001827029,0.0002523463,0.000006042664],"category_scores_gemma":[0.00005006491,0.0004337136,0.0001420663,0.0005178195,0.0003676476,0.000733813,0.000361414,0.0005703621,0.000001927101],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0004053269,"about_ca_system_score_gemma":0.0007588629,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000009189928,"about_ca_topic_score_gemma":0.000088552,"domain_scores_codex":[0.9960542,0.00002512398,0.0005158454,0.001718969,0.001034003,0.000651897],"domain_scores_gemma":[0.9981248,0.0002822076,0.0002480697,0.0009510687,0.0002379355,0.0001558984],"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":[0.00004449199,0.00002525286,0.00009931828,0.0002066901,0.00003762778,0.00003480405,0.001892456,0.1910803,0.00006253057,0.06227675,0.000002728585,0.744237],"study_design_scores_gemma":[0.0006401199,0.0002287097,0.00003984744,0.0007824539,0.00001766337,0.00005846133,9.191895e-7,0.9515245,0.002703031,0.04318726,0.0002062952,0.0006107443],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.0007007163,0.0003757262,0.9949391,0.0007079417,0.0007552087,0.0009475749,0.000002535624,0.0001320464,0.001439135],"genre_scores_gemma":[0.3822446,0.0000161091,0.6168576,0.0002980556,0.0002888553,0.00001902438,0.00001520908,0.00003294561,0.0002275683],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.7604442,"threshold_uncertainty_score":0.9998115,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.05235121322312605,"score_gpt":0.2453832477387255,"score_spread":0.1930320345155995,"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."}}