{"id":"W7106791207","doi":"10.48448/7zah-me62","title":"GRADE: Generating multi-hop QA and fine-gRAined Difficulty matrix for RAG Evaluation","year":2025,"lang":"","type":"other","venue":"Open MIND","topic":"","field":"","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Kootenay Association for Science & Technology; Discovery Air (Canada)","funders":"","keywords":"Inference; Construct (python library); Task (project management); Cluster analysis; Scalability; Key (lock); Semantics (computer science); Automated reasoning","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":[],"consensus_categories":[],"category_scores_codex":[0.007501399,0.002079767,0.001150711,0.004730845,0.0007097265,0.003039389,0.003078383,0.003033129,0.007269475],"category_scores_gemma":[0.05676688,0.0005116618,0.001268532,0.002083676,0.001197821,0.004317526,0.003777961,0.002303629,0.003059054],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001805526,"about_ca_system_score_gemma":0.001895299,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.007543453,"about_ca_topic_score_gemma":0.01052953,"domain_scores_codex":[0.9921793,0.003669952,0.0005938088,0.001627185,0.001635802,0.0002940991],"domain_scores_gemma":[0.9776881,0.01416838,0.001070433,0.004139645,0.002231723,0.0007017622],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"not_applicable","study_design_scores_codex":[0.002120507,0.001318048,0.02778681,0.002407579,0.0007089275,0.0006438001,0.001136873,0.2436598,0.013886,0.02152539,0.08897746,0.5958289],"study_design_scores_gemma":[0.0003734449,0.0005418191,0.005873055,0.0001408749,0.0001016251,0.000301882,0.0003291641,0.9377887,0.01154258,0.02862967,0.01426751,0.0001096982],"study_design_candidate":"not_applicable","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"other","genre_scores_codex":[0.1766029,0.003699928,0.7163345,0.001716013,0.0005968727,0.002452623,0.0213853,0.05963119,0.01758082],"genre_scores_gemma":[0.5599822,0.0003704775,0.4039013,0.0006003056,0.0001136529,0.001083678,0.02862423,0.001747016,0.003577104],"genre_candidate":"other","genre_consensus":null,"teacher_disagreement_score":0.007543453,"threshold_uncertainty_score":0.03967166,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1069548230903222,"score_gpt":0.4140181984159427,"score_spread":0.3070633753256204,"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."}}