{"id":"W4385570803","doi":"10.18653/v1/2023.findings-acl.187","title":"Automated Refugee Case Analysis: A NLP Pipeline for Supporting Legal Practitioners","year":2023,"lang":"en","type":"article","venue":"","topic":"Artificial Intelligence in Law","field":"Social Sciences","cited_by":7,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"Computer science; Named-entity recognition; Pipeline (software); Natural language processing; Artificial intelligence; Refugee; Transformer; Information extraction; Architecture; Domain (mathematical analysis); Matching (statistics); Test set; Deep learning; Information retrieval; Machine learning; Data mining; Law","routes":{"ca_aff":false,"ca_fund":false,"ca_venue":false,"about_ca":true,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":["insufficient_payload"],"consensus_categories":[],"category_scores_codex":[0.002153459,0.00009538705,0.0001843581,0.0002743528,0.000939502,0.0001743908,0.0001574121,0.0001019558,0.0009460849],"category_scores_gemma":[0.001995716,0.00009212259,0.0001829793,0.002575383,0.0001891221,0.000481618,0.00003371418,0.00008277807,0.0004003971],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00009432406,"about_ca_system_score_gemma":0.0002224268,"about_ca_topic_candidate":true,"about_ca_topic_consensus":true,"about_ca_topic_score_codex":0.01334885,"about_ca_topic_score_gemma":0.04783581,"domain_scores_codex":[0.9983355,0.0001239535,0.000434422,0.000288391,0.0003101442,0.000507581],"domain_scores_gemma":[0.9987244,0.0005058886,0.0001810042,0.0001778044,0.000268041,0.0001428628],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00008862541,0.0002208186,0.002186324,0.00003752514,0.0007548257,0.002354146,0.02382366,0.009767129,0.0009342469,0.7100378,0.2256681,0.02412683],"study_design_scores_gemma":[0.00009782978,0.00004134408,0.00004815127,0.000007942093,0.0004628268,0.00002857544,0.06756511,0.6106899,0.001615911,0.003762482,0.3153212,0.0003587134],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.7550353,0.00001780163,0.07015956,0.02576552,0.001471231,0.001897007,0.0001149993,0.0108738,0.1346648],"genre_scores_gemma":[0.9815406,0.0000052694,0.001917238,0.0002728626,0.0002832719,0.00006939034,0.0000579066,0.00001350809,0.01583997],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.7062753,"threshold_uncertainty_score":0.9999672,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.07342167803979849,"score_gpt":0.4534971779005597,"score_spread":0.3800754998607612,"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."}}