OrcaLoca: An LLM Agent Framework for Software Issue Localization

Zhongming Yu, Hejia Zhang, Yujie Zhao, Hanxian Huang, Matrix Yao, Ke Ding, Jishen Zhao
Proceedings of the 42nd International Conference on Machine Learning, PMLR 267:73416-73436, 2025.

Abstract

Recent developments in Large Language Model (LLM) agents are revolutionizing Autonomous Software Engineering (ASE), enabling automated coding, problem fixes, and feature improvements. However, localization – precisely identifying software problems by navigating to relevant code sections – remains a significant challenge. Current approaches often yield suboptimal results due to a lack of effective integration between LLM agents and precise code search mechanisms. This paper introduces OrcaLoca, an LLM agent framework that improves accuracy for software issue localization by integrating priority-based scheduling for LLM-guided action, action decomposition with relevance scoring, and distance-aware context pruning. Experimental results demonstrate that OrcaLoca becomes the new open-source state-of-the-art (SOTA) in function match rate (65.33%) on SWE-bench Lite. It also improves the final resolved rate of an open-source framework by 6.33 percentage points through its patch generation integration.

Cite this Paper

BibTeX
@InProceedings{pmlr-v267-yu25x, title = {{O}rca{L}oca: An {LLM} Agent Framework for Software Issue Localization}, author = {Yu, Zhongming and Zhang, Hejia and Zhao, Yujie and Huang, Hanxian and Yao, Matrix and Ding, Ke and Zhao, Jishen}, booktitle = {Proceedings of the 42nd International Conference on Machine Learning}, pages = {73416--73436}, year = {2025}, editor = {Singh, Aarti and Fazel, Maryam and Hsu, Daniel and Lacoste-Julien, Simon and Berkenkamp, Felix and Maharaj, Tegan and Wagstaff, Kiri and Zhu, Jerry}, volume = {267}, series = {Proceedings of Machine Learning Research}, month = {13--19 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v267/main/assets/yu25x/yu25x.pdf}, url = {https://proceedings.mlr.press/v267/yu25x.html}, abstract = {Recent developments in Large Language Model (LLM) agents are revolutionizing Autonomous Software Engineering (ASE), enabling automated coding, problem fixes, and feature improvements. However, localization – precisely identifying software problems by navigating to relevant code sections – remains a significant challenge. Current approaches often yield suboptimal results due to a lack of effective integration between LLM agents and precise code search mechanisms. This paper introduces OrcaLoca, an LLM agent framework that improves accuracy for software issue localization by integrating priority-based scheduling for LLM-guided action, action decomposition with relevance scoring, and distance-aware context pruning. Experimental results demonstrate that OrcaLoca becomes the new open-source state-of-the-art (SOTA) in function match rate (65.33%) on SWE-bench Lite. It also improves the final resolved rate of an open-source framework by 6.33 percentage points through its patch generation integration.} }
Endnote
%0 Conference Paper %T OrcaLoca: An LLM Agent Framework for Software Issue Localization %A Zhongming Yu %A Hejia Zhang %A Yujie Zhao %A Hanxian Huang %A Matrix Yao %A Ke Ding %A Jishen Zhao %B Proceedings of the 42nd International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2025 %E Aarti Singh %E Maryam Fazel %E Daniel Hsu %E Simon Lacoste-Julien %E Felix Berkenkamp %E Tegan Maharaj %E Kiri Wagstaff %E Jerry Zhu %F pmlr-v267-yu25x %I PMLR %P 73416--73436 %U https://proceedings.mlr.press/v267/yu25x.html %V 267 %X Recent developments in Large Language Model (LLM) agents are revolutionizing Autonomous Software Engineering (ASE), enabling automated coding, problem fixes, and feature improvements. However, localization – precisely identifying software problems by navigating to relevant code sections – remains a significant challenge. Current approaches often yield suboptimal results due to a lack of effective integration between LLM agents and precise code search mechanisms. This paper introduces OrcaLoca, an LLM agent framework that improves accuracy for software issue localization by integrating priority-based scheduling for LLM-guided action, action decomposition with relevance scoring, and distance-aware context pruning. Experimental results demonstrate that OrcaLoca becomes the new open-source state-of-the-art (SOTA) in function match rate (65.33%) on SWE-bench Lite. It also improves the final resolved rate of an open-source framework by 6.33 percentage points through its patch generation integration.
APA
Yu, Z., Zhang, H., Zhao, Y., Huang, H., Yao, M., Ding, K. & Zhao, J.. (2025). OrcaLoca: An LLM Agent Framework for Software Issue Localization. Proceedings of the 42nd International Conference on Machine Learning, in Proceedings of Machine Learning Research 267:73416-73436 Available from https://proceedings.mlr.press/v267/yu25x.html.

Related Material