List your company

Legal Question Answering on DISC-Law-SFT (test)

90.07Win RateLLM-AutoDP
82.99884.83486.6788.506Jan 28, 2026

Evaluation results

MethodLinks
LLM-AutoDPBase Model=Qwen2.5-7B, Judge=GPT-42026.0190.071.458.48
LLM-AutoDPBase Model=Gemma-2-9B, Judge=GPT-42026.0189.410.2410.35
LLM-AutoDPBase Model=Llama3.1-8B, Judge=LawLLM2026.0186.643.949.42
LLM-AutoDPBase Model=Llama3.1-8B, Judge=GPT-42026.0184.66015.34
LLM-AutoDPBase Model=Qwen2.5-7B, Judge=LawLLM2026.0184.196.259.56
LLM-AutoDPBase Model=Gemma-2-9B, Judge=LawLLM2026.0183.2710.496.24