UI-Venus Technical Report: Building High-performance UI Agents with RFT

TL;DR

UI-Venus is a high-performance UI agent using multimodal large language models and reinforcement fine-tuning, achieving state-of-the-art results in UI grounding and navigation tasks with only screenshots as input.

Contribution

The paper introduces UI-Venus, a novel UI agent leveraging RFT and a multimodal LLM, with new techniques for data cleaning and trajectory refinement to enhance performance.

Findings

UI-Venus achieves SOTA on UI grounding benchmarks.

UI-Venus outperforms existing models in UI navigation success rate.

Proposed methods improve reasoning and planning in complex UI tasks.

Abstract

We present UI-Venus, a native UI agent that takes only screenshots as input based on a multimodal large language model. UI-Venus achieves SOTA performance on both UI grounding and navigation tasks using only several hundred thousand high-quality training samples through reinforcement finetune (RFT) based on Qwen2.5-VL. Specifically, the 7B and 72B variants of UI-Venus obtain 94.1% / 50.8% and 95.3% / 61.9% on the standard grounding benchmarks, i.e., Screenspot-V2 / Pro, surpassing the previous SOTA baselines including open-source GTA1 and closed-source UI-TARS-1.5. To show UI-Venus's summary and planing ability, we also evaluate it on the AndroidWorld, an online UI navigation arena, on which our 7B and 72B variants achieve 49.1% and 65.9% success rate, also beating existing models. To achieve this, we introduce carefully designed reward functions for both UI grounding and navigation tasks and corresponding efficient data cleaning strategies. To further boost navigation performance, we propose Self-Evolving Trajectory History Alignment & Sparse Action Enhancement that refine historical reasoning traces and balances the distribution of sparse but critical actions, leading to more coherent planning and better generalization in complex UI tasks. Our contributions include the publish of SOTA open-source UI agents, comprehensive data cleaning protocols and a novel self-evolving framework for improving navigation performance, which encourage further research and development in the community. Code is available at https://github.com/inclusionAI/UI-Venus.