UI-Voyager: A Self-Evolving GUI Agent Learning via Failed Experience

TL;DR

UI-Voyager is a self-evolving GUI agent that improves learning efficiency from failed experiences using a two-stage process, achieving high success rates in mobile GUI automation without manual data annotation.

Contribution

The paper introduces UI-Voyager, a novel two-stage self-evolving GUI agent leveraging Rejection Fine-Tuning and Group Relative Self-Distillation for improved learning from failures.

Findings

Achieves 81.0% Pass@1 success rate on AndroidWorld

Outperforms recent baselines and exceeds human-level performance

Effectiveness of GRSD verified through ablation and case studies

Abstract

Autonomous mobile GUI agents have attracted increasing attention along with the advancement of Multimodal Large Language Models (MLLMs). However, existing methods still suffer from inefficient learning from failed trajectories and ambiguous credit assignment under sparse rewards for long-horizon GUI tasks. To that end, we propose UI-Voyager, a novel two-stage self-evolving mobile GUI agent. In the first stage, we employ Rejection Fine-Tuning (RFT), which enables the continuous co-evolution of data and models in a fully autonomous loop. The second stage introduces Group Relative Self-Distillation (GRSD), which identifies critical fork points in group rollouts and constructs dense step-level supervision from successful trajectories to correct failed ones. Extensive experiments on AndroidWorld show that our 4B model achieves an 81.0% Pass@1 success rate, outperforming numerous recent baselines and exceeding human-level performance. Ablation and case studies further verify the effectiveness of GRSD. Our method represents a significant leap toward efficient, self-evolving, and high-performance mobile GUI automation without expensive manual data annotation.