Paper Detail

SWE-AGILE: A Software Agent Framework for Efficiently Managing Dynamic Reasoning Context

Shuquan Lian, Juncheng Liu, Yazhe Chen, Yuhong Chen, Hui Li

huggingface Score 9.5

Published 2026-04-13 · First seen 2026-04-14

General AI

Abstract

Prior representative ReAct-style approaches in autonomous Software Engineering (SWE) typically lack the explicit System-2 reasoning required for deep analysis and handling complex edge cases. While recent reasoning models demonstrate the potential of extended Chain-of-Thought (CoT), applying them to the multi-turn SWE task creates a fundamental dilemma: retaining full reasoning history leads to context explosion and ``Lost-in-the-Middle'' degradation, while discarding it would force the agent to redundantly re-reason at every step. To address these challenges, 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. Empirically, SWE-AGILE sets a new standard for 7B-8B models on SWE-Bench-Verified using only 2.2k trajectories and 896 tasks. Code is available at https://github.com/KDEGroup/SWE-AGILE.

Workflow Status

Review status
pending
Role
unreviewed
Read priority
now
Vote
Not set.
Saved
no
Collections
Not filed yet.
Next action
Not filled yet.

Reading Brief

No structured notes yet. Add `summary_sections`, `why_relevant`, `claim_impact`, or `next_action` in `papers.jsonl` to enrich this view.

Why It Surfaced

No ranking explanation is available yet.

Tags

No tags.

BibTeX

@misc{lian2026swe,
  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},
  abstract = {Prior representative ReAct-style approaches in autonomous Software Engineering (SWE) typically lack the explicit System-2 reasoning required for deep analysis and handling complex edge cases. While recent reasoning models demonstrate the potential of extended Chain-of-Thought (CoT), applying them to the multi-turn SWE task creates a fundamental dilemma: retaining full reasoning history leads to context explosion and ``Lost-in-the-Middle'' degradation, while discarding it would force the agent to},
  url = {https://huggingface.co/papers/2604.11716},
  keywords = {ReAct-style approaches, System-2 reasoning, Chain-of-Thought, multi-turn SWE task, context explosion, Lost-in-the-Middle degradation, dynamic reasoning context, sliding window, reasoning digests, SWE-Bench-Verified, code available, huggingface daily},
  eprint = {2604.11716},
  archiveprefix = {arXiv},
}

Metadata

{}