COBALT: Contextual Bandit Learning with Offline Trajectories for Multi-Turn Code Generation
arXiv:2602.03806v2 Announce Type: replace-cross Abstract: Training large language models (LLMs) for multi-turn code generation typically relies on online reinforcement learning (RL), in which models learn from repeatedly generated online rollouts. Although effective, online RL can be costly and unstable. In this paper, we present contextual bandit learning with offline trajectories (COBALT), a simple method that reuses offline multi-turn trajectories to efficiently train LLMs for multi-turn cod…