List your company

Concept Alignment on RepE (test)

100HARMSelf Modulation
-4235077Jan 2, 2025

Evaluation results

MethodLinks
Self ModulationTarget LLM (mt)=Llama3.1, Source LLM (ms)=Llama3.12025.01100368.749.28
Self ModulationTarget LLM (mt)=Llama2, Source LLM (ms)=Llama22025.0196568.527.26
L-Cross ModulationTarget LLM (mt)=Llama2, Source LLM (ms)=Qwen22025.0196649.168.84
L-Cross ModulationTarget LLM (mt)=Llama2, Source LLM (ms)=Llama3.12025.0196548.927.96
Self ModulationTarget LLM (mt)=Qwen2, Source LLM (ms)=Qwen22025.0188507.046.26
No ModulationTarget LLM (mt)=Llama2, Source LLM (ms)=None2025.010985.565.74