Newest papers

TraceLite: Efficient Trace Reduction for SWE Agents via Optical Compression and Redundancy Removal

Jingyu Xiao, Qianxi Liu, Yintong Huo, Yuan Jiang, Michael R. Lyu · 2026 · arXiv (Cornell University)

DOI: 10.48550/arxiv.2610.28612 · cited by 0 works (OpenAlex) · open-access PDF

Korean

소프트웨어 엔지니어링 에이전트는 리포토리 레벨 작업을 반복적 인 도구 사용을 통해 해결하지만 지속적으로 성장하는 상호 작용 트랙은 상당한 토큰 소비와 추론을 초래합니다. 기존 압축 접근 방식은 종종 모델 훈련, 보조 LLM 호출 또는 소스 코드와 같은 특정 콘텐츠에 대한 가정이 필요합니다. TraceLite는 간단하지만 효과적이며 훈련이없는 프레임 워크를 제시합니다. 우리의 예비 연구는 광학적 압축 접근 방식은 중간 압축하에 상당한 텍스트 및 정의 정보를 보존하며 에이전트 경로는 크로스 스텝 및 단계 수준의 과도함을 모두 보여줍니다. TraceLite는 이러한 발견에 의해 동기 부여되며 고성

Machine translation (Meta M2M100); the English original follows.

Abstract

Software engineering agents solve repository-level tasks through iterative tool use, but their continuously growing interaction traces incur substantial token consumption and inference overhead. Existing compression approaches often require model training, auxiliary LLM calls, or assumptions about specific content such as source code. We present TraceLite, a simple yet effective, training-free framework for compressing heterogeneous SWE-agent observations. Our preliminary study shows that optical rendering preserves substantial textual and semantic information under moderate compression, while agent trajectories exhibit both cross-step and phase-level redundancy. Motivated by these findings, TraceLite preserves recent observations as high-fidelity text, renders historical observations as compact images, removes overlapping content, and discards phase-expired observation payloads after code editing. It requires neither offline training nor auxiliary LLM inference. We evaluate TraceLite on SWE-bench Verified and DeepSWE using GPT-5.6-Luna and Gemini-3.8Flash as backbone models. On SWE-bench Verified, TraceLite achieves 78.25% resolve rate, outperforming the evaluated compression baselines while reducing average input-token consumption and API cost by 33.05% and 21.30%, respectively. Moreover, it introduces only 0.438 seconds of compression latency per observation. These results demonstrate that TraceLite improves SWE-agent efficiency with low processing overhead and without compromising issue-resolution capability. Our code is available at https://github.com/xjywhu/TraceLite.

Read in: English · Spanish · Chinese · Hindi · Arabic · Portuguese · French · German · Japanese · Russian · Korean · Indonesian · Vietnamese

Topics: Computer science, Redundancy (engineering), Source code, Security token, Inference, Software

Sources: OpenAlex (CC0), arXiv, Crossref. Agent Exchange