ANCHOR: Branch-Point Data Generation for GUI Agents

TL;DR

ANCHOR is a framework that efficiently expands GUI interaction data from limited seed demonstrations by generating diverse, goal-consistent trajectories for training desktop agents.

Contribution

It introduces a scalable trajectory expansion method using branch points and verification to improve GUI agent training data quality and diversity.

Findings

Models trained on expanded data outperform zero-shot baselines.

ANCHOR-generated data improves generalization across applications.

The framework achieves better performance on desktop benchmarks.

Abstract

End-to-end GUI agents for real desktop environments require large amounts of high-quality interaction data, yet collecting human demonstrations is expensive and existing synthetic pipelines often suffer from limited task diversity or noisy, goal-drifting trajectories. We present a trajectory expansion framework Anchor that bootstraps scalable desktop supervision from a small set of verified seed demonstrations. Starting from each seed, we identify branch points that correspond to meaningful state changes and propose new, state-grounded task variants conditioned on the current GUI context. An executing agent then follows the proposed instructions to generate new trajectories, while a verifier enforces task completion via state-aware checks and trajectory-level consistency. To improve supervision quality, we further apply task-conditioned step-level filtering to remove ungrounded actions and denoise post-branch segments to maintain coherent intent. Experiments on standard desktop benchmarks, OSWorld and WindowsAgentArena, show that models fine-tuned on our expanded corpus achieve consistent improvements over zero-shot agents and representative synthesis baselines, and generalize across applications and operating systems.