{"id":"W4416037239","doi":"10.18653/v1/2025.emnlp-main.6","title":"QFrCoLA: a Quebec-French Corpus of Linguistic Acceptability Judgments","year":2025,"lang":"","type":"article","venue":"","topic":"Natural Language Processing Techniques","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Université Laval","funders":"Fonds de recherche du Québec – Nature et technologies; Natural Sciences and Engineering Research Council of Canada","keywords":"Corpus linguistics; Set (abstract data type); Term (time); Subject (documents)","routes":{"ca_aff":true,"ca_fund":true,"ca_venue":false,"about_ca":true,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001547299,0.001475052,0.0004066839,0.002693589,0.002466299,0.001435704,0.001549484,0.001493398,0.0122478],"category_scores_gemma":[0.01056961,0.0002697604,0.0005182489,0.002016645,0.001104271,0.001059328,0.001281624,0.001951832,0.004485636],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00608797,"about_ca_system_score_gemma":0.004771116,"about_ca_topic_candidate":true,"about_ca_topic_consensus":true,"about_ca_topic_score_codex":0.6275851,"about_ca_topic_score_gemma":0.7772655,"domain_scores_codex":[0.9977412,0.0007114934,0.0001217077,0.0004791633,0.0007786843,0.000167715],"domain_scores_gemma":[0.9934874,0.001984868,0.0002365686,0.000770661,0.003206352,0.000314245],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"not_applicable","study_design_gemma":"not_applicable","study_design_scores_codex":[0.0007770937,0.0005755813,0.02805109,0.001600166,0.0002152098,0.001243269,0.004588186,0.006141217,0.01521186,0.00699332,0.7470951,0.1875079],"study_design_scores_gemma":[0.0003338958,0.0002400852,0.2029814,0.0006379941,0.0001228791,0.001235286,0.006655212,0.04894818,0.01485454,0.004533101,0.7190596,0.0003978868],"study_design_candidate":"not_applicable","study_design_consensus":"not_applicable","genre_codex":"dataset","genre_gemma":"dataset","genre_scores_codex":[0.4178775,0.003520324,0.02244575,0.003665447,0.000733751,0.00110302,0.463331,0.01437636,0.07294691],"genre_scores_gemma":[0.4588008,0.0005994663,0.02853575,0.0008572385,0.0001237005,0.0007881444,0.4911934,0.001157204,0.01794434],"genre_candidate":"dataset","genre_consensus":"dataset","teacher_disagreement_score":0.3724149,"threshold_uncertainty_score":0.7492163,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01136993307768044,"score_gpt":0.2944397419745004,"score_spread":0.28306980889682,"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."}}