{"ID":2875459,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2509.02360","arxiv_id":"2509.02360","title":"When Agents go Astray: Course-Correcting SWE Agents with PRMs","abstract":"Large Language Model (LLM) agents are increasingly deployed for complex, multi-step software engineering (SWE) tasks. However, their trajectories often contain costly inefficiencies, such as redundant exploration, looping, and failure to terminate once a solution is reached. Prior work has largely treated these errors in a post-hoc manner, diagnosing failures only after execution. In this paper, we introduce SWE-PRM, an inference-time Process Reward Model (PRM) that intervenes during execution to detect and course-correct trajectory-level errors. Our PRM design leverages a taxonomy of common inefficiencies and delivers lightweight, interpretable feedback without modifying the underlying policy. On SWE-bench Verified, closed-source PRMs improve resolution from 40.0% to 50.6% (+10.6 p.p.), with the largest gains on medium and hard tasks. Among feedback strategies, taxonomy-guided PRMs outperform unguided or explicit action-prescriptive variants, increasing success rate while reducing trajectory length. These benefits come at an acceptable added inference cost of as low as $0.2, making PRMs a practical and scalable mechanism for improving SWE agents' reliability and efficiency.","short_abstract":"Large Language Model (LLM) agents are increasingly deployed for complex, multi-step software engineering (SWE) tasks. However, their trajectories often contain costly inefficiencies, such as redundant exploration, looping, and failure to terminate once a solution is reached. Prior work has largely treated these errors...","url_abs":"https://arxiv.org/abs/2509.02360","url_pdf":"https://arxiv.org/pdf/2509.02360v2","authors":"[\"Shubham Gandhi\",\"Jason Tsay\",\"Jatin Ganhotra\",\"Kiran Kate\",\"Yara Rizk\"]","published":"2025-09-02T14:23:15Z","proceeding":"cs.AI","tasks":"[\"cs.AI\",\"cs.SE\"]","methods":"[\"Large Language Model\",\"Language Model\",\"LoRA\"]","has_code":false}