{"id":"W4281263937","doi":"10.18653/v1/2022.findings-acl.117","title":"Two-Step Question Retrieval for Open-Domain QA","year":2022,"lang":"en","type":"article","venue":"Findings of the Association for Computational Linguistics: ACL 2022","topic":"Topic Modeling","field":"Computer Science","cited_by":3,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Ministry of Science and ICT, South Korea; National Research Foundation of Korea; National Research Foundation","keywords":"Search engine indexing; Inference; Computer science; Information retrieval; Labrador Retriever; Pipeline (software); Squid; Domain (mathematical analysis); Artificial intelligence; Open domain; Question answering; Mathematics; Programming language","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.003484777,0.000825689,0.001031021,0.001333484,0.0006761005,0.001198632,0.002568392,0.001790962,0.009398755],"category_scores_gemma":[0.01089002,0.0006647364,0.00122987,0.0009476984,0.0009235206,0.005374018,0.002892502,0.002800946,0.005467695],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001233987,"about_ca_system_score_gemma":0.002147533,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.008352006,"about_ca_topic_score_gemma":0.01053769,"domain_scores_codex":[0.9985671,0.0005922961,0.00008777731,0.000401242,0.0002341387,0.0001175059],"domain_scores_gemma":[0.9948623,0.002682291,0.0001662064,0.001366412,0.0007102555,0.000212625],"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":[0.001002704,0.0008574103,0.005205947,0.0009548726,0.0001828752,0.0002082218,0.0008622739,0.07719907,0.02542557,0.02889133,0.03443011,0.8247796],"study_design_scores_gemma":[0.00008736369,0.0001965959,0.0008150071,0.00002186545,0.00003672318,0.0001967949,0.00007878431,0.9617288,0.01028136,0.02015907,0.006366363,0.0000312849],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.02775576,0.001041913,0.9518597,0.0008017536,0.0001089852,0.0003724749,0.0006811127,0.01422962,0.003148724],"genre_scores_gemma":[0.4850205,0.0004853074,0.5036844,0.0005585844,0.0001637943,0.0003594253,0.003324152,0.000459717,0.005944135],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.009398755,"threshold_uncertainty_score":0.03144199,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01875750815666395,"score_gpt":0.2925474942045004,"score_spread":0.2737899860478364,"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."}}