Newest papers

Back Up Before You Regret: Measuring Backup Awareness and Automating Backup in Coding Agents

Xu Li, Simon Yu, Minzhou Pan, Yiyou Sun, Zhaorun Chen, Bo Li, Weiyan Shi · 2026 · arXiv (Cornell University)

DOI: 10.48550/arxiv.2610.28934 · cited by 0 works (OpenAlex) · open-access PDF

Portuguese

Os agentes de codificação estão agora completando tarefas de forma autónoma, modificando o estado do mundo real, como artefatos, ambientes de software e configurações ao longo do caminho. No entanto, eles mostram consciência de backup limitada: reconhecendo que uma ação irá modificar o estado que pode ser necessário mais tarde, e preservando-o antecipadamente. Para avaliar a consciência de backup, introduzimos o RegretBench, um benchmark de 205 tarefas executáveis baseadas em momentos de arrependimento relatados em plataformas como X.com, GitHub e Stack Overflow, onde os usuários explicitamente se arrependeram de não manter o estado de custo que precisavam mais tarde. Em seis modelos de fronteira, os agentes completam 86,1% das tarefas em média, mas prescrevendo ações antes desses

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

Abstract

Coding agents now complete tasks autonomously, modifying real-world state such as artifacts, software environments, and configurations along the way. Yet they show limited backup awareness: recognizing that an action will modify state that may later be needed, and preserving it beforehand. To evaluate backup awareness, we introduce RegretBench, a benchmark of 205 executable tasks grounded in regret moments reported on platforms such as X.com, GitHub, and Stack Overflow, where users explicitly regretted not preserving state they later needed. Across six frontier models, agents complete 86.1% of tasks on average yet back up before these moments in only 15.2%, showing that task completion ability does not imply backup awareness. Because it is hard to predict which changes will be regretted, we propose BackUpper, a harness plug-in backup layer that checks every tool call and preserves affected state before modification. It combines rule-based strategies for common operations with a lightweight subagent for complex cases, and its rule table can expand with subagent decisions over time. On RegretBench, BackUpper covers every regret moment action and restores states with 92.2% average similarity to the original. Achieving this costs only $0.014 per task, with 7x fewer tokens than the main-agent self-backup baseline. When users have preferences on what to back up, BackUpper can also narrow its scope accordingly, matching human annotations on whether an action is worth preserving with an 84.5% F1 score. These results suggest that agents can keep actions reversible at low cost while respecting users' preferences.

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

Topics: Backup, Computer science, Oracle, Leverage (statistics), Coding (social sciences), Regret

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