Newest papers

Over-Defense in Coding Agents Begins at Silent Forks

Hanjun Luo, Xuecheng Liu, Xiucheng Zhang, Zhuoning Xu, Yingbin Jin, Yankai Chen, Xue Liu, Hanan Salam · 2026 · arXiv (Cornell University)

DOI: 10.48550/arxiv.2610.24949 · cited by 0 works (OpenAlex) · open-access PDF

Portuguese

Os modelos de linguagem por trás dos agentes de codificação são pós-treinados com recompensas de resultado, e os métodos recentes concentram crédito nas voltas que determinam o resultado. Descobrimos que as ações defensivas além do que uma tarefa requer escapam ambos. Eles começam em forks silenciosos, as voltas onde cada ação seguinte leva ao mesmo resultado. Tais ações deixam a taxa de sucesso inalterada, e modelos de fronteira levam o ramo desproporcional lá em até 91% das ações decisivas sob orçamentos implícitos. A correção mais direta, um checker na trajetória de recompensa, atinge essas forcas apenas através de trajetórias totalmente proporcionais, deixando 74% dos grupos de GRPO com recompensas idênticas e comportamento inalterado. Isto motiva um repensar onde os sina

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

Abstract

Language models behind coding agents are post-trained with outcome rewards, and recent methods concentrate credit on the turns that decide the outcome. We find that defensive actions beyond what a task requires escape both. They begin at silent forks, the turns where every next action leads to the same outcome. Such actions leave the success rate unchanged, and frontier models take the disproportionate branch there in up to 91% of decisive actions under implicit budgets. The most direct fix, a checker in the trajectory reward, reaches these forks only through fully proportionate trajectories, leaving 74% of GRPO groups with identical rewards and behavior unchanged. This motivates a rethink of where learning signals belong, namely where behavior diverges, which a behavioral checker can locate and the outcome cannot. Our method, ForkDPO, contrasts a proportionate and a disproportionate action at the same prefix, at silent forks on both sides of the required behavior. On a test set of unseen repositories, it lowers a 4B model's disproportionate behavior from 70.3% to 58.9%, against 67.7% for DPO at randomly chosen turns, establishing behavioral divergence as a learning signal for how agents act beyond whether they finish. The code is publicly available at https://github.com/Astarojth/silent-forks.

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

Topics: Computer science, Outcome (game theory), Coding (social sciences), Task (project management), Set (abstract data type), Action (physics)

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