List your company

Medical Question Answering Robustness on medical testbed 800-question

76.88IDCR-FT
67.904870.234972.56574.8951Apr 23, 2026

Evaluation results

MethodLinks
R-FTModel Backbone=Llama-3.1-8B2026.0476.8899.84
RBED+R-FTModel Backbone=Llama-3.1-8B2026.0476.8899.87
PBTModel Backbone=Llama-3.1-8B2026.0474.561.4
DuET-PDModel Backbone=Llama-3.1-8B2026.0474.3811.23
Warning PromptModel Backbone=Llama-3.1-8B2026.0468.251.38
VanillaModel Backbone=Llama-3.1-8B2026.0468.251.55
RBEDModel Backbone=Llama-3.1-8B2026.0468.258