Recovering Policy-Induced Errors: Benchmarking and Trajectory Synthesis for Robust GUI Agents
:::info Stub — Full Engineering Breakdown Coming This paper has a linked code implementation and was featured on Hugging Face Papers with 14 upvotes. A full breakdown with production viability rating, implementation notes, and honest limitations is being written. Subscribe to AI Letters → :::
| Authors | Tianpeng Bu et al. |
| Year | 2026 |
| HF Upvotes | 14 |
| arXiv | 2605.29447 |
| Download | |
| Code | https://github.com/AlibabaResearch/RoTS |
Abstract
While GUI agents have advanced rapidly, they often lack the robustness to recover from their own errors, hindering real-world deployment. To bridge this gap at both the evaluation and data levels, we introduce GUI-RobustEval and propose Robustness-driven Trajectory Synthesis. GUI-RobustEval contains 1,216 executable test cases that systematically measure error recovery capabilities across a broad and realistic spectrum of error modes. At the data level, RoTS is a scalable synthesis framework that creates 800k high-quality data via a tree-based pipeline that proactively discovers diverse error modes and synthesizes corresponding recovery steps. Our two models, RoTS-7B and RoTS-32B, fine-tuned on our dataset, both demonstrate significant gains on GUI-RobustEval and traditional GUI benchmarks. Notably, RoTS-32B achieves state-of-the-art performance on OSWorld, with a 47.4% success rate and a 33.8% All-Pass@4 score, suggesting that improved long-horizon error recovery ability contributes to both robustness and overall performance. Our code is available at https://github.com/AlibabaResearch/RoTS.
Engineering Breakdown
The Problem
While GUI agents have advanced rapidly, they often lack the robustness to recover from their own errors, hindering real-world deployment.
The Approach
To bridge this gap at both the evaluation and data levels, we introduce GUI-RobustEval and propose Robustness-driven Trajectory Synthesis.
Key Results
Notably, RoTS-32B achieves state-of-the-art performance on OSWorld, with a 47.4% success rate and a 33.8% All-Pass@4 score, suggesting that improved long-horizon error recovery ability contributes to both robustness and overall performance.
Research Areas
This paper contributes to the following areas of AI/ML engineering:
- Machine learning
- Deep learning
- Neural networks
- Model optimization
- AI systems
- Recovering
:::tip Subscribe Get weekly breakdowns of papers like this in AI Letters - the newsletter for engineers building production AI systems. :::