Coding agents fix more bugs by choosing whole files without compression
Skip the Compressor: Coding Agents Fix More Issues When They Pick Whole Files
Software Engineering
Summary
When AI coding helpers try to fix software bugs, they can only look at a limited amount of code at once. Many methods try to pick parts of code to show by cutting or ranking pieces, but this can miss important changes or alter the code in ways that make fixes harder. The authors show that picking entire files exactly as they are, using a simple selection method called CQ2-CARE, helps these AI helpers fix more issues than other ways. This approach works better because it shows the AI the exact code it needs to edit, without missing files or changing code.
What this means in practice
- •For software development teams: Improve automated bug fixing tools by selecting whole files exactly to increase repair success on large codebases.
- •For software testing teams: Use file-level selection methods in AI-assisted debugging to create more accurate patches that pass validation tests more often.
Authors
Xiaocong Yang, Rishit Chatterjee, Himanshu Udupi, Vikram Ramavarapu, ChengXiang Zhai
Abstract
A coding agent works on repositories that can be far larger than its context window can hold. Prompt compressors and retrievers offer solutions to this problem, but patching an issue needs two things at once: the files that must change, and their code exactly as written, since the model must reproduce the lines it edits. Retrievers rank code by its similarity to the issue rather than by whether it must change, and compressors delete text to fit the budget. We show that at an equal token budget, an agent resolves more issues when the budget goes to choosing which files to show verbatim, as our simple selector CQ2-CARE does, than when it goes to pruning tokens or retrieving chunks. We fit every context to the same per-issue budget on 300 recent SWE-rebench issues and grade each patch with the benchmark's tests. With GPT-5.6 and Claude Opus 5, CQ2-CARE resolves 43.7% and 46.7% of the issues, 6.3 points more than the best of three compressors with either model, and 9.3 and 12.7 points more than the best of three retrievers. CQ2-CARE even resolves more issues than two methods trained for issue resolution, SweRank (by 0.7 and 1.7 points) and SWE-Pruner (by 2.0 and 7.3). Our analysis shows the defects of the baseline methods: retrievers often miss the files to edit, and compressors keep them but alter their text, so fewer patches apply. Exact-file selection needs no compression model and shows every line as the repository has it, so the model can quote the code it edits.