4 papers · 1 filter
Reasoning as Gradient: Scaling MLE Agents Beyond Tree Search
Yifei Zhang, Xu Yang, Xiao Yang +8
LLM-based agents for machine learning engineering (MLE) predominantly rely on tree search, a form of gradient-free optimization that uses scalar validation scores to rank candidate…
MarkovScale: Towards Optimal Sequential Scaling at Inference Time
Youkang Wang, Jian Wang, Rubing Chen +3
Sequential scaling is a prominent inference-time scaling paradigm, yet its performance improvements are typically modest and not well understood, largely due to the prevalence of h…
OptScale: Probabilistic Optimality for Inference-time Scaling
Youkang Wang, Jian Wang, Rubing Chen +1
Inference-time scaling has emerged as a powerful technique for enhancing the reasoning performance of Large Language Models (LLMs). However, existing approaches often rely on heuri…
OptPO: Optimal Rollout Allocation for Test-time Policy Optimization
Youkang Wang, Jian Wang, Rubing Chen +3
Test-time policy optimization enables large language models (LLMs) to adapt to distribution shifts by leveraging feedback from self-generated rollouts. However, existing methods re…