paper-with-me

홈 › Papers

ScreenExplorer: Training a Vision-Language Model for Diverse Exploration in Open GUI World

2025-05-25 · Runliang Niu, Jinglong Ji, Yi Chang, Qi Wang

The rapid progress of large language models (LLMs) has sparked growing interest in building Artificial General Intelligence (AGI) within Graphical User Interface (GUI) environments. However, existing GUI agents based on LLMs or vision-language models (VLMs) often fail to generalize to novel environments and rely heavily on manually curated, diverse datasets. To overcome these limitations, we introduce ScreenExplorer, a VLM trained via Group Relative Policy Optimization(GRPO) in real, dynamic, and open-ended GUI environments. Innovatively, we introduced a world-model-based curiosity reward function to help the agent overcome the cold-start phase of exploration. Additionally, distilling experience streams further enhances the model's exploration capabilities. Our training framework enhances model exploration in open GUI environments, with trained models showing better environmental adaptation and sustained exploration compared to static deployment models. Our findings offer a scalable pathway toward AGI systems with self-improving capabilities in complex interactive settings.

📄 PDF Abstract BibTeX arXiv:2505.19095

Code (1)

niuzaisheng/screenexplorer 공식 구현 pytorch

Tasks

Language ModelingLanguage Modelling

Similar Papers 제목 키워드 기반

Exploration with Principles for Diverse AI Supervision

2023-10-13 · Hao liu, Matei Zaharia, Pieter Abbeel

Training large transformers using next-token prediction has given rise to groundbreaking advancements in AI. While this generative AI approach has produced impressive results, it heavily leans on human supervision. Even …

Reinforcement Learning (RL)Unsupervised Reinforcement Learning

CuriousBot: Interactive Mobile Exploration via Actionable 3D Relational Object Graph

2025-01-23 · YiXuan Wang, Leonor Fermoselle, Tarik Kelestemur, Jiuguang Wang 외

Mobile exploration is a longstanding challenge in robotics, yet current methods primarily focus on active perception instead of active interaction, limiting the robot's ability to interact with and fully explore its envi…

Object

VLN-Zero: Rapid Exploration and Cache-Enabled Neurosymbolic Vision-Language Planning for Zero-Shot Transfer in Robot Navigation

2025-09-23 · Neel P. Bhatt, Yunhao Yang, Rohan Siva, Pranay Samineni 외 arxiv

Rapid adaptation in unseen environments is essential for scalable real-world autonomy, yet existing approaches rely on exhaustive exploration or rigid navigation policies that fail to generalize. We present VLN-Zero, a t…

Vision-Language NavigationRobot Navigation

ActiveVLN: Towards Active Exploration via Multi-Turn RL in Vision-and-Language Navigation

2025-09-16 · Zekai Zhang, Weiye Zhu, Hewei Pan, Xiangchen Wang 외 arxiv

The Vision-and-Language Navigation (VLN) task requires an agent to follow natural language instructions and navigate through complex environments. Existing MLLM-based VLN methods primarily rely on imitation learning (IL)…

Reinforcement Learning

Visual-Language Navigation Pretraining via Prompt-based Environmental Self-exploration

2021-11-16 · ACL ARR November 2021 11 · Anonymous

Vision-language navigation (VLN) is a challenging task due to its large searching space in the environment. To address this problem, previous works have proposed some methods of fine-tuning a large model that pretrained …

Domain AdaptationVision-Language Navigation