arXiv · 2410.02089
RLEF: Grounding Code LLMs in Execution Feedback with Reinforcement Learning
Abstract
Large language models (LLMs) deployed as agents solve user-specified tasks over multiple steps while keeping the required manual engagement to a minimum. Crucially, such LLMs need to ground their generations in any feedback obtained to reliably achieve the desired outcomes. We propose an end-to-end reinforcement learning method for teaching models to leverage execution feedback in the realm of code synthesis, where state-of-the-art LLMs struggle to improve code iteratively compared to independent sampling. We benchmark on competitive programming tasks, where we achieve new state-of-the art results with both small (8B parameters) and large (70B) models while reducing the amount of samples required by an order of magnitude. Our analysis of inference-time behavior demonstrates that our method produces LLMs that effectively leverage automatic feedback over multiple steps.
Explore related subjects
Keep this discovery
Explore connections, maps & timelines
Jonas Gehring, Kunhao Zheng, Jade Copet, Vegard Mella, Quentin Carbonneaux, Taco Cohen, Gabriel Synnaeve. 2024-10-02. RLEF: Grounding Code LLMs in Execution Feedback with Reinforcement Learning. https://arxiv.org/abs/2410.02089
Cite the original work for its findings. Save a collection to share your selection of sources.