PAUSE is a steering framework for coding agents that intervenes once before the agent’s first source‑code edit, eliminating the need for learned monitors, checkpoint searches, or extra model calls. By having a Planner generate a repair plan from the task and pre‑edit transcript, the same agent can resume execution with this plan, preserving context and workspace. Experiments on SWE‑bench Verified and Multilingual show that PAUSE improves resolution rates by up to 16.8% online and recovers 35.4% of failed tasks offline, while reducing Executor token usage by 33‑41% compared to restarts.
By Viral Patel, Boyin Tan, Jason Xue, Youcheng Sun
The paper studies when to intervene in large language model (LLM) agents that interact with web services. Using the ReInBench dataset of 1,259 agent traces and 1,403 error events, the authors find that 725 errors self‑recover, often within a few steps, and that self‑recovery can aid subsequent error handling and overall task completion. They propose ReCoF, a predictor that decides whether an error will self‑recover, achieving a 9.02% relative improvement in Macro‑F1 over the best baseline.
By Chiming Duan, Tong Jia, Yingxi Chen, Minghua He, Lingzhe Zhang, Ying Li
The paper introduces Traverse, a benchmark of 2,518 agent trajectories and 6,967 annotated mistakes across software engineering, computer use, and science tasks, revealing that failures often go unrecovered and can cause irreversible harm before a run is deemed successful. It shows that human judges struggle to detect the first mistake in most runs, while a 4‑billion‑parameter verifier called Scout can locate failures more effectively and improve task success when used to select among candidate runs. The study demonstrates that making failure detection inexpensive and reliable can enable long‑horizon agents to learn from their own mistakes and increase trustworthiness in autonomous AI.
By Salman Rahman, Yubin Kim, Mihir Parmar, A. Ali Heydari, Genglin Liu, Simon A. Lee, Weizhi Zhang, Arian Hosseini, Ahmed A. Metwally, Yuzhe Yang, Baharan Mirzasoleiman, Xin Liu, Pavel Izmailov, Saadia Gabriel, Mark Malhotra, Shwetak Patel, Daniel McDuff, Hamid Palangi
arXiv:2608.24569v2 Announce Type: replace
Abstract: Large language model (LLM) agents build on one another's work through summaries, plans, and other intermediate artifacts. These records let a downs...
By Yiheng Sun, Huifei Wang, Yancheng Zhu, Zhenyu Li, Zebin Zhao, Yifan Yuan
The paper treats human interventions during LLM agent failures as labels that pinpoint the internal computation that went wrong, rather than merely training data. In a controlled environment, the authors analyze one failure type—over‑asking a clarifying question when the request is already clear—using rule‑based overseer labels validated by blind human judgment. They find that the decision to ask versus act can be causally localized across several models, that over‑asking exhibits distinct mechanistic signatures, and that steering or weight consolidation can mitigate the issue, though outcomes vary by model family.
By Cesare Zavattari, Alessandro Tommasi, Giuseppe Prencipe
Software engineering (SWE) agents resolve repository-level issues through long trajectories that grow increasingly expensive as context accumulates. Failed runs tend to be longer and exhibit redundant exploration or looping, suggesting that some failures may be detectable before completion.
The paper introduces CTXGAP, a study measuring how much information coding agents omit from their user-facing reports compared to their full execution logs. Across 37 coding tasks, agents answer 93.6% of task-state questions when asked directly, but report-only readers recover only 52.0% of the same information, revealing a 40.7‑point gap. Even at final checkpoints of verified fixes, the gap remains significant, and detailed terminal scrollback improves recovery only to 73.6%. The study highlights that missing details in reports can leave supervisors unaware of failures, workarounds, and risks that need scrutiny.
By Guoyang Xu, Rui Zhang, Xiuyu Wu, Hao Chen
arXiv:2610.14146v1 Announce Type: new
Abstract: Long-horizon terminal benchmarks measure whether an agent can sustain progress through extended workflows with process rewards, not only whether its fi...
By Zongxia Li, Zhongzhi Li, Yucheng Shi, Ruhan Wang, Junyao Yang, Chengsong Huang, Zhichao Liu, Xiyang Wu, Anhao Li, Yue Yu, Ergan Shang, Fuxiao Liu, Tong Zheng, Ninghao Liu, Lichao Sun, Haitao Mi, Leowei Liang
LEASH is a training‑free framework that enables uncertainty‑aware escalation from a weak to a strong computer‑use agent. It uses latent uncertainty signals from the weak agent’s hidden states to decide when to hand off control, and a shadow observer to maintain oversight of the strong black‑box agent. Experiments on OSWorld show LEASH improves task success over standalone weak agents and outperforms random escalation, achieving comparable performance to Claude Opus 5 while cutting inference cost by 19.2%.
By Qingni Wang, Changdae Oh, Yuzhe Yang, Sharon Li, Jiachen Yang, Ang Li, Xin Eric Wang
AgentEval is a system that tracks a contract-bound assessment record throughout a stateful agent’s tool execution, capturing state, events, and evidence to determine whether obligations are satisfied, violated, or unresolved. In 32 tool‑world administrations, independent checks confirmed all 30 complete episodes, while controlled interventions on 48 evidence views uncovered 36 binding faults that hash‑only checks missed. A public‑benchmark study on 192 held‑out executions identified ten completed subgoals within broader task failures and highlighted the impact of missing evidence on overall task success.
By Ruiqi Gao (Alibaba Cloud, Alibaba Group)
arXiv:2608. 03222v1 Announce Type: cross Abstract: Software engineering (SWE) agents resolve repository-level issues through long trajectories that grow increasingly expensive as context accumulates.
By Chenyu Wang, Yunbo Lyu, Junda He, Zhou Yang, Chenxing Zhong, Yaniv Harel, David Lo
arXiv:2603.01209v3 Announce Type: replace
Abstract: In CodeAct, language-model agents write Python that calls tools and use execution feedback to choose actions. Persistent runtimes preserve Python v...
By Victor May, Van Khue Nguyen, Aaditya Salgarkar, Yishan Wang, Diganta Misra, Huu Nguyen