1 paper · 2 filters
Shuyu Gan, James Mooney, Pan Hao +4
Large Language Model (LLM) agents can increasingly automate complex reasoning through Test-Time Scaling (TTS), iterative refinement guided by reward signals. However, many real-wor…