arXiv Machine Learning
Critical-State RL: Diagnosing Trainable States for Multi-Turn Tool Use
Read the original on arXiv Machine Learning →The Flow has not summarised this story yet — read it at arXiv Machine Learning.
Related stories
arXiv AI
Oct 15
DeltaTTG: Training Multi-Turn Agents with a Token-to-Go Progress Meter
arXiv:2610.17666v1 Announce Type: cross Abstract: Reinforcement learning with verifiable rewards (RLVR) for multi-turn agents relies on sparse terminal rewards that say little about individual steps....
arXiv AI
Jul 28
Progress-conditioned Group Policy Optimization for Long-Horizon Agentic Tasks
arXiv:2607. 22724v1 Announce Type: cross Abstract: Group-based policy optimization has been increasingly used to train large language model (LLM) agents from sparse outcome rewards by comparing trajectories or steps within a group.
arXiv AI
Jul 13
Prompt-Driven Exploration
arXiv:2607. 08837v1 Announce Type: cross Abstract: Exploration is essential to RL since a policy cannot improve by repeatedly sampling the behaviors it already prefers.
arXiv AI
Jun 2
On Effectiveness and Efficiency of Agentic Tool-calling and RL Training
arXiv:2606. 00135v1 Announce Type: cross Abstract: Tool-calling is a central component of modern large language model (LLM) agents, equipping them with skills beyond their parametric knowledge.
arXiv AI
Sep 21
Prompt-Driven Exploration: Language as an Exploration Space for VLA Reinforcement Learning
arXiv:2607.08837v4 Announce Type: replace-cross Abstract: Exploration is essential to RL since a policy cannot improve by repeatedly sampling the behaviors it already prefers. Standard methods inject...
arXiv AI
Oct 12
COBALT: Contextual Bandit Learning with Offline Trajectories for Multi-Turn Code Generation
COBALT is a contextual bandit learning method that reuses offline multi-turn code generation trajectories to train large language models efficiently. It treats multi-turn code generation as a one-step recoverable Markov decision process, collecting partial trajectories from a reference LLM and training the target model to complete each partial trajectory in a single step. Experiments on LiveCodeBench show that COBALT improves Pass@1 scores for R1-Distill 8B and Qwen3 8B by up to 9.0 and 6.2 points, matches the performance of two online RL baselines, and enhances robustness against misleading execution feedback.