{"id":"W3205430877","doi":"10.1007/978-3-030-88483-3_23","title":"Towards Unifying the Explainability Evaluation Methods for NLP","year":2021,"lang":"en","type":"book-chapter","venue":"Lecture notes in computer science","topic":"Topic Modeling","field":"Computer Science","cited_by":1,"is_retracted":false,"has_abstract":false,"ca_institutions":"University of Ottawa","funders":"","keywords":"Interpretability; Computer science; Artificial intelligence; Phrase; Metric (unit); Set (abstract data type); Machine learning; Feature (linguistics); Focus (optics); Natural language; Deep learning; Key (lock); Natural language processing; Training set","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":["metaresearch"],"consensus_categories":[],"category_scores_codex":[0.03422286,0.002010187,0.003004182,0.01042464,0.001219432,0.01007649,0.004227939,0.003076725,0.00575785],"category_scores_gemma":[0.08928218,0.001313635,0.003950854,0.006019047,0.003442884,0.01556414,0.007569923,0.006494556,0.002106216],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.002515267,"about_ca_system_score_gemma":0.003044933,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.003794084,"about_ca_topic_score_gemma":0.004195668,"domain_scores_codex":[0.9721746,0.0138109,0.003215113,0.002729207,0.007384778,0.0006854265],"domain_scores_gemma":[0.8934941,0.07566746,0.003789636,0.01335276,0.01242729,0.00126878],"domain_codex":null,"domain_gemma":"evaluation","domain_candidate":"evaluation","domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"theoretical_or_conceptual","study_design_scores_codex":[0.0002383315,0.0002464073,0.004146778,0.001065375,0.0004033434,0.00009485707,0.001286015,0.01620567,0.003669543,0.3310704,0.005830172,0.6357432],"study_design_scores_gemma":[0.00004996031,0.0001336174,0.00256807,0.0005173329,0.0002604426,0.0001366754,0.0004997077,0.3376995,0.004696964,0.6334565,0.01985884,0.0001223061],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.002002978,0.0008826519,0.9936463,0.0004928162,0.0000738663,0.0001432049,0.0001940748,0.0008938565,0.001670418],"genre_scores_gemma":[0.06470568,0.000843628,0.9296021,0.0002254918,0.0002932489,0.0004866624,0.001138891,0.0006953687,0.002008914],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.9657772,"threshold_uncertainty_score":0.1809899,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.08998709768257471,"score_gpt":0.3780216545665107,"score_spread":0.288034556883936,"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."}}