List your company

Persona-based Role-Playing Faithfulness Evaluation on Complicated Famous Figures (test)

0.9Spartacus ΔAPCEU
0.7921.5212.252.979May 13, 2024

Evaluation results

MethodLinks
EUModel Family=Gemma-7B, PRP Strategy=Experience Uploading, APC-based DPO=false2024.050.9-1.1-5.5-3.2-1.66.8-0.7
EU + APC-based DPOModel Family=Gemma-7B, PRP Strategy=Experience Uploading, APC-based DPO=true2024.052.20.8-0.7-0.2-1.36.90.2
Gemma-7BModel Family=Gemma-7B, PRP Strategy=Vanilla, APC-based DPO=false2024.052.32.31.42.43.59.63.5
GPT-4Model Family=GPT, PRP Strategy=Vanilla, APC-based DPO=false2024.052.52.5-21.55.115.14.1
ChatGPTModel Family=GPT, PRP Strategy=Vanilla, APC-based DPO=false2024.052.61.4-3-0.61.711.92.3
Gemma-7B + APC-based DPOModel Family=Gemma-7B, PRP Strategy=Vanilla, APC-based DPO=true2024.052.93.24.823.118.15.6
RAG + APC-based DPOModel Family=Gemma-7B, PRP Strategy=Retrieval-Augmented Generation, APC-based DPO=true2024.053.43.9536.419.96.9
RAGModel Family=Gemma-7B, PRP Strategy=Retrieval-Augmented Generation, APC-based DPO=false2024.053.6333.45.416.35.7