Results show that how information is represented to an agent shapes both its cost and its success, making the environment worth optimizing in addition to the parameters and the harness, making the environment worth optimizing in addition to the parameters and the harness.
Ming-Xuan Li, Huan-Zhi Mao, Qiu-Yang Mang et al.· 0 citations
This work introduces Stateless Language Agents (SLAs), built on the principle of stateful search with stateless agents, and evaluates SLA against three recent frameworks on software engineering, kernel optimization, and algorithm design at budgets of up to one billion tokens.
Qizheng Zhang, Chang Ji, I. Sun et al.· 2 citations
A key strength of Proximal Policy Optimization (PPO) is its learned critic, which uses historical trajectories collected during reinforcement learning to estimate expected returns and reduce policy-gradient variance. However, we find that the critic is also a major source of instability in reinforcement learning for la...
Xuan-Yi Zhou, Qiu-Yang Mang, Huan-Zhi Mao et al.· arXiv.org· 0 citations
Combee is proposed, a novel framework to scale parallel prompt learning for self-improving agents that speeds up learning and enables running many agents in parallel while learning from their aggregate traces without quality degradation.
This work defines the scaling inflection point as the per-session budget where marginal Elo gains match the independent-sampling reference, and proposes Elo-per-token analysis, which tracks the best solution found at each token budget and uses a Bradley-Terry model to aggregate within-task orderings into Elo ratings ac...