Model

Gemma4-12B

A large language model in the Gemma series released by Google DeepMind, with a parameter scale of 12B.

Timeline (last 90 days)

Oct 30

On the Qwen3-14B and Gemma4-12B backbones, GUIDER outperforms CRPO, MRPO and FaithMed on the out-of-domain benchmarks AMEGA-LLM, RealMedQA and CancerGUIDE.

Unverified50%

All Facts (1)

Unverified

On the Qwen3-14B and Gemma4-12B backbones, GUIDER outperforms CRPO, MRPO and FaithMed on the out-of-domain benchmarks AMEGA-LLM, RealMedQA and CancerGUIDE.

50%

Source Articles

GUIDER: Bridging the Path–Answer Gap in Clinical Guideline ReasoningOct 30