RLEF: Grounding Code LLMs in Execution Feedback with Reinforcement Learning
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 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 start-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.
Code (0)
등록된 구현이 없습니다.
Tasks
reinforcement-learningReinforcement LearningSimilar Papers 제목 키워드 기반
HyCodePolicy: Hybrid Language Controllers for Multimodal Monitoring and Decision in Embodied Agents
Recent advances in multimodal large language models (MLLMs) have enabled richer perceptual grounding for code policy generation in embodied agents. However, most existing systems lack effective mechanisms to adaptively m…
Multimodal ReasoningRobot ManipulationProgram SynthesisMRT at IberLEF-2025 PRESTA Task: Maximizing Recovery from Tables with Multiple Steps
This paper presents our approach for the IberLEF 2025 Task PRESTA: Preguntas y Respuestas sobre Tablas en Español (Questions and Answers about Tables in Spanish). Our solution obtains answers to the questions by implemen…
Code GenerationOctopus: Embodied Vision-Language Programmer from Environmental Feedback
Large vision-language models (VLMs) have achieved substantial progress in multimodal perception and reasoning. When integrated into an embodied agent, existing embodied VLM works either output detailed action sequences a…
BenchmarkingCode GenerationDecision MakingMinecraftCo-RedTeam: Orchestrated Security Discovery and Exploitation with LLM Agents
Large language models (LLMs) have shown promise in assisting cybersecurity tasks, yet existing approaches struggle with automatic vulnerability discovery and exploitation due to limited interaction, weak execution ground…
Vulnerability DetectionTowards Execution-Grounded Automated AI Research
Automated AI research holds great potential to accelerate scientific discovery. However, current LLMs often generate plausible-looking but ineffective ideas. Execution grounding may help, but it is unclear whether automa…
Reinforcement Learning