Large language model (LLM) agents built on the ReAct pattern concentrate four responsibilities in one model: selecting a strategy, choosing each action, formatting it, and judging whether the result is adequate. Nothing outside the generative loop can reject its output, so an agent that cannot progress does not report...
A pilot implementing the acceptance- and unit-level verifier pairs is reported, comparing five configurations that share one executor, tool set, and scorer and differ only in verification, on the four-hop stratum of MuSiQue with an 8B-parameter backbone.
Temperature-Scaled On-Policy Self-Distillation (TS-OPSD), a lightweight policy reheating method that internalizes the exploratory effect of temperature into model parameters, suggests that entropy restoration can serve as a simple post-collapse intervention for extending reasoning-oriented RL.
Xue Yang, Jiachen Yu, J. Wu et al.· arXiv.org· 1 citation
AssumptionMiner is presented, a framework that makes implicit assumptions a first-class artifact of LLM-based code generation, and demonstrates that making assumptions explicit improves the transparency and controllability of LLM-based code generation.
This work proposes AgentOPSD, a critic-free, recursive method for turn-level credit assignment in agentic reinforcement learning that aggregates token-level teacher-student log-probability gaps into turn-level evidence and recursively updates a Bayesian belief state in log-odds space.
Zi-Han Wang, Zhengxi Lu, Zhiyuan Yao et al.· 7 citations· ⚡1
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
