List your company

Long-context Summary on Summary Task (test)

63.2Multi-turn Coverage ScoreRLSTA
34.80842.17949.5556.921Mar 5, 2026

Evaluation results

MethodLinks
RLSTAModel=Qwen3-4B Instruct-25072026.0363.275.883.5
BaseModel=Qwen3-4B Instruct-25072026.0355.57574
BaseModel=Qwen2.5-7B Instruct2026.0344.659.475.1
RLSTAModel=Qwen2.5-7B Instruct2026.0344.15974.7
RLSTAModel=Llama-3.2-3B Instruct2026.0343.455.578.2
RLSTAModel=Qwen2.5-3B Instruct2026.0343.35283.2
BaseModel=Llama-3.2-3B Instruct2026.0338.454.969.8
BaseModel=Qwen2.5-3B Instruct2026.0335.952.468.5