Viral Patel, Boyin Tan, Jason Xue, Youcheng Sun · 2026 · arXiv
DOI: 10.48550/arXiv.2610.29870 · open-access PDF
Agen pengkodean dapat menginvestasikan upaya substansial dalam eksplorasi dan pengujian repository namun masih menghasilkan patch yang tidak berhasil. Metode pengendalian online yang ada meninjau trajektori secara berkala atau memanggil penasihat ketika monitor mendeteksi masalah perilaku; ulasan sering mahal, dan kurangnya perilaku anomali tidak berarti bahwa agen berada di jalur yang benar. Kami mengusulkan PAUSE, kerangka kerja untuk mengarahkan agen pengkodean dengan intervensi perencanaan tunggal. agen mencapai editing kode sumber pertama mereka pada median 60% dari kemajuan trajektori, setelah eksplorasi, reproduksi bug, dan pengujian.
Machine translation (Meta M2M100); the English original follows.
Coding agents can invest substantial effort in repository exploration and testing yet still produce unsuccessful patches. Existing online steering methods review trajectories periodically or invoke an advisor when a monitor detects behavioral problems; frequent reviews are costly, and the absence of anomalous behavior does not mean that an agent is on the right track. We propose PAUSE, a framework for steering coding agents with a single planning intervention. Agents reach their first source-code edit at a median of 60% of trajectory progress, after exploration, bug reproduction, and testing.
Read in: English · Spanish · Chinese · Hindi · Arabic · Portuguese · French · German · Japanese · Russian · Korean · Indonesian · Vietnamese
GET https://store.agentexchange.work/gw/papers/verify?title=...&year=... (0.003 USD), fetch this record as JSON with /gw/papers/paper?id=2610.29870, or list what it cites with /gw/papers/references?id=.... Pay per call in USDC (x402) or with a 19 USD card key for 1,000 calls. Catalogue · llms.txtSources: OpenAlex (CC0), arXiv, Crossref. Agent Exchange