--- tags: - agent pipeline_tag: text-generation library_name: transformers --- # SWE-AGILE ## 📣 News [2026/02/23] SWE-AGILE has been accepted to the ACL 2026 Findings.
[[📖 Paper](https://huggingface.co/papers/2604.11716)] [[🤗 Checkpoints](https://huggingface.co/KDEGroup)] [[🤗 Daily Paper](https://huggingface.co/papers/2604.11716)] [[🚀 Github](https://github.com/KDEGroup/SWE-AGILE)]
 ## 🔥 Overview Prior approaches typically lack the **explicit System-2 reasoning** required for deep analysis. While recent reasoning models demonstrate the potential of extended Chain-of-Thought (CoT), applying them to multi-turn tasks creates a dilemma: retaining full history leads to **context explosion**, while discarding it causes **redundant re-reasoning**. We propose SWE-AGILE, a novel software agent framework designed to bridge the gap between reasoning depth, efficiency, and context constraints. SWE-AGILE introduces a Dynamic Reasoning Context strategy, maintaining a “sliding window” of detailed reasoning for immediate continuity to prevent redundant re-analyzing, while compressing historical reasoning content into concise Reasoning Digests via **Backfilling Data Synthesis**, **Trajectory Snapshot Training** and **Compression-Aware Optimization**. While our current paradigm implicitly reduces redundant state reconstruction, a highly promising direction to strictly enforce this efficiency is to quantitatively monitor the reasoning content. By calculating the embedding similarity between consecutive reasoning steps or employing an LLM-as-a-Judge, future iterations can explicitly filter out repetitive SFT trajectories or design targeted RLVR penalties, pushing the boundary of cognitive efficiency even further. ![overview](https://cdn-uploads.huggingface.co/production/uploads/669f53549a21428ccda89fab/YQmyhamGa29mmAWAh1Qoe.png) ![swe-bench-verified](https://cdn-uploads.huggingface.co/production/uploads/669f53549a21428ccda89fab/YGIov1OkGhAcE4wsPBCs2.png) ## ⭐️ Citation If you find this project useful, please cite our work: ```bibtex @misc{lian2026sweagilesoftwareagentframework, title={SWE-AGILE: A Software Agent Framework for Efficiently Managing Dynamic Reasoning Context}, author={Shuquan Lian and Juncheng Liu and Yazhe Chen and Yuhong Chen and Hui Li}, year={2026}, eprint={2604.11716}, archivePrefix={arXiv}, primaryClass={cs.AI}, url={https://arxiv.org/abs/2604.11716}, } ``` ## 🤝 Acknowledgements We sincerely thank the projects [R2E-Gym/R2E-Gym](https://github.com/R2E-Gym/R2E-Gym) and [rllm-org/rllm](https://github.com/rllm-org/rllm) for providing their open-source resources.