Skip to content

Locating Hidden Failures Makes Long-Horizon Agents More Reliable

Sep 2026 · 2 citations · 59 references
Computer Science

TL;DR

By making failure cheap to locate and correct, this work is a foundation for more trustworthy long-horizon agents that learn from their own mistakes, and a practical path to overseeing increasingly autonomous AI.

Abstract

As AI agents take on long, autonomous tasks, we increasingly oversee rather than perform the work, yet we still judge them almost entirely by whether they finally succeed. An outcome cannot reveal where a run went wrong, whether the agent recovered, or the irreversible harm it caused along the way, and where long-horizon agents fail remains unmapped. We study $2518$ agent trajectories across software engineering, computer use, and science, close to real deployment, and classify $6967$ mistakes into $78$ failure types. Failure follows a recurring signature: after its first mistake an agent often fails to recover and rarely catches the error itself, so the run continues unchecked while still looking correct; whether an agent recovers depends on the task and the environment's feedback, not on the agent framework running it. Long-horizon agents can do real harm on the way to a passing result: even runs scored as solved delete data, corrupt systems, or fabricate success rather than earning it. We release these human-verified annotations as Traverse, a benchmark on which six frontier judges struggle to locate failure regardless of scale: even the strongest correctly identifies the first mistake in fewer than a third of runs. Yet Scout, a $4$B verifier we trained, locates failure far better than these judges and transfers to domains it never saw. Used at test time to select among an agent's candidate runs, it raises task success above the agent's own single-attempt performance, without retraining the agent. By making failure cheap to locate and correct, this work is a foundation for more trustworthy long-horizon agents that learn from their own mistakes, and a practical path to overseeing increasingly autonomous AI.

View source

Similar papers

#artificial intelligence Preprint Oct 2026

Agents Are Systems, Not Models: Rethinking Agentic Evaluation

Agent evaluations should be evaluated as configurable systems themselves, and that some desired behaviors are more effectively implemented in the system than requested through prompting, according to a new benchmark of four scientific tasks.

Luis Wiedmann, Leander Girrbach, Cordelia Schmid et al. · 0 citations
Preprint Sep 2026

The Missing Boundary: How Autonomous Agents Lose Control

Autonomous agents increasingly perform long-horizon tasks involving tool use, persistent state, and consequential actions, raising a fundamental question: \emph{under what conditions does an agent cross the boundary of authorized execution while pursuing a legitimate task?} Existing studies often attribute such failure...

Zong-Hao Ying, Xiang-Fan Wu, Bo Yang et al. · 1 citation · ⚡1

Staying on Task: Testing the Foundations of Long-Horizon Agent Reliability

Long-horizon agentic workflows require models to sustain repeated state-dependent actions all while the context grows, sub-task complexity changes, and new data arrives. Each situation represents an independent axis along which an agent may fail. An agent reconciling a long ledger, for example, must repeatedly read its...

Jeffrey Willette, Krishna C. Puvvada, Boris Ginsburg · 0 citations

Same Winners, Different Success Rates: Evaluating How LLM Agents Recover from Failures

It is proved that certifying exact population ties is impossible in finite time, and that the assignment of outcomes to checkpoints carries information beyond marginal outcome distributions, and proposed four diagnostic quantities that expose this failure mode with no additional data collection.

Dong Xu, Zhang-Fan Yang, Jian-Tao Wu et al. · 0 citations

When Successful Strategies Fail: Adaptation to Environmental Novelty in Terminal Agents

This work introduces AGNI, an automated pipeline that extracts trajectory-relevant assumptions, injects targeted environmental changes, and validates that the resulting novel tasks remain solvable and highlights a gap between task competence and adaptive capability and motivate environmental variation as a core dimensi...

Janvijay Singh, Vaishnavi Shrivastava, Dilek Hakkani-Tur et al. · 0 citations

Related blog posts

Microsoft Research Blog Oct 12, 2026

Simulating student learning for better AI tutoring

StudentSim is a research system for evaluating and improving AI tutoring tools before they are tested with learners. It models how individuals may engage with tutors, giving feedback to improve tutor performance across subjects like chess, writing, and mathematics. The post Simulating student learning for better AI tutoring appeared first on Microsoft Research.

Microsoft Research Blog Nov 16, 2026

TokenEconomizer: Learning from Experience to Make AI Coding Agents More Efficient

AI coding agents are increasingly help developers fix bugs and build features largely on their own, but every interaction with their models consumes tokens. In each session, an agent sends instructions, tool descriptions, and session history to its model. Most efforts to reduce this cost focus on individual tasks. Yet developers return to the same […] The post TokenEconomizer: Learning from Experience to Make AI Coding Agents More Efficient appeared first on Microsoft Research.

Microsoft Research Blog Nov 12, 2026

Microsoft Research Focus – Edition #98

What's new in research at Microsoft? Host Anson Ho explores the latest, including AI-powered universal translation, a breakthrough in a 31-year-old mathematics challenge, and advances making AI coding agents faster and more efficient.  The post Microsoft Research Focus – Edition #98 appeared first on Microsoft Research.

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.