Katalog Plus
Bibliothek der Frankfurt UAS
Bald neuer Katalog: sichern Sie sich schon vorab Ihre persönlichen Merklisten im Nutzerkonto: Anleitung.
Dieses Ergebnis aus BASE kann Gästen nicht angezeigt werden.  Login für vollen Zugriff.

Training Long-Context, Multi-Turn Software Engineering Agents with Reinforcement Learning

Title: Training Long-Context, Multi-Turn Software Engineering Agents with Reinforcement Learning
Authors: Golubev, Alexander; Trofimova, Maria; Polezhaev, Sergei; Badertdinov, Ibragim; Nekrashevich, Maksim; Shevtsov, Anton; Karasik, Simon; Abramov, Sergey; Andriushchenko, Andrei; Fisin, Filipp; Skvortsov, Sergei; Yangel, Boris
Publication Year: 2025
Collection: ArXiv.org (Cornell University Library)
Subject Terms: Machine Learning; Computation and Language; Software Engineering
Description: Research on applications of reinforcement learning (RL) to large language models has mostly been focused on single-turn problems, such as mathematical reasoning or single-shot code generation. While these problems can be viewed as token-level multi-turn Markov decision processes (MDPs), this view corresponds to a degenerate case of multi-turn interaction where the environment provides no feedback. This contrasts with many real-world domains, such as software engineering (SWE), which require rich multi-turn interactions with a stateful environment that responds to each action with a non-trivial observation. To bridge this gap, we demonstrate the successful application of RL to this general regime. Our methodology begins with rejection fine-tuning (RFT) using execution feedback to train a policy to follow instructions and formatting effectively, followed by a synchronous RL pipeline using DAPO for iterative improvement. Applying this pipeline to Qwen2.5-72B-Instruct, we increase its Pass@1 on the SWE-bench Verified benchmark from 11% to 39%, substantially improving upon the 20% RFT baseline. On the May and June splits of SWE-rebench, the resulting agent achieves Pass@1 of 35% and 31% respectively, competitive with even larger models such as DeepSeek-V3-0324 or Qwen3-235B-A22B, demonstrating that our methodology offers a practical approach for training capable agents for multi-turn interactive tasks using open-weight models.
Document Type: text
Language: unknown
Relation: http://arxiv.org/abs/2508.03501
Availability: http://arxiv.org/abs/2508.03501
Accession Number: edsbas.CD455B41
Database: BASE