{"id":"W3214151748","doi":"10.18653/v1/2021.mrl-1.12","title":"Mr. TyDi: A Multi-lingual Benchmark for Dense Retrieval","year":2021,"lang":"en","type":"article","venue":"","topic":"Topic Modeling","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Natural Sciences and Engineering Research Council of Canada; Canada First Research Excellence Fund; Compute Canada","keywords":"Computer science; Benchmark (surveying); Ranking (information retrieval); Relevance (law); Artificial intelligence; Natural language processing; Information retrieval; Representation (politics); Point (geometry); Resource (disambiguation); Machine learning; Mathematics; Geography","routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"about_ca":false,"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":[0.005776885,0.00296801,0.002089771,0.007847357,0.002723857,0.00355911,0.004143484,0.002985316,0.01220317],"category_scores_gemma":[0.01920391,0.0005887406,0.002122662,0.008265987,0.001510122,0.005421742,0.004668533,0.002387833,0.012881],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.002570363,"about_ca_system_score_gemma":0.003521397,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.04696874,"about_ca_topic_score_gemma":0.07484346,"domain_scores_codex":[0.9929646,0.002082522,0.0009170637,0.001278556,0.002063337,0.000693954],"domain_scores_gemma":[0.9911901,0.002366711,0.0005181071,0.002871865,0.002374223,0.0006790726],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"not_applicable","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.001461351,0.001457493,0.00630651,0.004622561,0.0007308591,0.0005715133,0.0005389742,0.0138623,0.01179249,0.005309647,0.7420002,0.211346],"study_design_scores_gemma":[0.001619588,0.002583578,0.03240307,0.0007265173,0.0006859622,0.0042994,0.002857279,0.2130751,0.03782503,0.01405185,0.6891237,0.0007488686],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"dataset","genre_gemma":"empirical","genre_scores_codex":[0.2158812,0.02418074,0.0613436,0.004873117,0.004103779,0.003651427,0.5728925,0.04501484,0.06805868],"genre_scores_gemma":[0.1196916,0.001668824,0.06696429,0.0009908493,0.0004750916,0.001153769,0.7935824,0.001405349,0.0140678],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.04696874,"threshold_uncertainty_score":0.0933907,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.05023543945653591,"score_gpt":0.3006741292895153,"score_spread":0.2504386898329793,"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."}}