一作ACL 2026Findings· Findings of ACL
Towards Inference-time Scaling for Continuous Space Reasoning
Minghan Wang · Thuy-Trang Vu · Ehsan Shareghi · Gholamreza Haffari
摘要
Inference-time scaling through multiple sample generation in combination with Process- or Outcome-Reward Model (PRM or ORM) re-ranking has proven effective for text-based reasoning in large language models. This paper investigates whether such established techniques can be successfully adapted to...