List your company
ResearchFormerly WizWand

Counterfactual Recommendation on Medical Communication Responses

99.38Improvement RateGreedy Recommendations
44.88459.03273.1887.328Jun 17, 2026
Updated 4mo ago

Evaluation results

MethodLinks
Greedy RecommendationsAuditor=EBM, Extractor=Gemma-4-31B2026.0699.383.6
Greedy RecommendationsAuditor=EBM, Extractor=Gemma-4-31B (R)2026.0699.235
Counterfactual RecommendationsAuditor=EBM, Extractor=Gemma-4-31B (R)2026.0698.916.6
Greedy RecommendationsAuditor=EBM, Extractor=Qwen3.5-27B (R)2026.0698.184.3
Counterfactual RecommendationsAuditor=EBM, Extractor=Qwen3.5-27B (R)2026.0697.546.3
Counterfactual RecommendationsAuditor=EBM, Extractor=Qwen3.5-27B2026.0697.196.3
Counterfactual RecommendationsAuditor=LR, Extractor=Gemma-4-31B (R)2026.0695.847
Counterfactual RecommendationsAuditor=EBM, Extractor=Gemma-4-31B2026.0694.916.4
Greedy RecommendationsAuditor=EBM, Extractor=Qwen3.5-27B2026.0694.784.1
Counterfactual RecommendationsAuditor=LR, Extractor=Gemma-4-31B2026.0694.727.6
Counterfactual RecommendationsAuditor=Average across auditors and extractors2026.0693.186.41
Counterfactual RecommendationsAuditor=EBM, Extractor=Qwen3-80B2026.0689.585.3
Counterfactual RecommendationsAuditor=LR, Extractor=Qwen3-80B2026.0689.555.3
Greedy RecommendationsAuditor=EBM, Extractor=Qwen3-80B2026.0689.493.4
Counterfactual RecommendationsAuditor=LR, Extractor=Qwen3.5-27B (R)2026.0688.686.8
Greedy RecommendationsAuditor=LR, Extractor=Qwen3-80B2026.0688.292.2
Greedy RecommendationsAuditor=LR, Extractor=Gemma-4-31B (R)2026.0688.232.3
Counterfactual RecommendationsAuditor=LR, Extractor=Qwen3.5-27B2026.0686.266.5
Greedy RecommendationsAuditor=Average across auditors and extractors2026.0685.142.95
Greedy RecommendationsAuditor=LR, Extractor=Qwen3.5-27B (R)2026.0674.942
Greedy RecommendationsAuditor=LR, Extractor=Qwen3.5-27B2026.0671.951.6
Greedy RecommendationsAuditor=LR, Extractor=Gemma-4-31B2026.0646.981