VAGEN: Reinforcing World Model Reasoning for Multi-Turn VLM Agents

• URL: http://arxiv.org/abs/2510.16907v1
• Date: Sun, 19 Oct 2025 16:05:07 GMT
• Title: VAGEN: Reinforcing World Model Reasoning for Multi-Turn VLM Agents
• Authors: Kangrui Wang, Pingyue Zhang, Zihan Wang, Yaning Gao, Linjie Li, Qineng Wang, Hanyang Chen, Chi Wan, Yiping Lu, Zhengyuan Yang, Lijuan Wang, Ranjay Krishna, Jiajun Wu, Li Fei-Fei, Yejin Choi, Manling Li,
• Abstract summary: Key challenge in training Vision-Language Model (VLM) agents, compared to Language Model (LLM) agents, is shift from textual states to complex visual observations.<n>We ask: Can VLM agents construct internal world models through explicit visual state reasoning?<n>We architecturally enforce and reward the agent's reasoning process via reinforcement learning (RL)<n>We find that the agent's reasoning into State Estimation and Transition Modeling is critical for success.
• Score: 130.70999337445468
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: A key challenge in training Vision-Language Model (VLM) agents, compared to Language Model (LLM) agents, lies in the shift from textual states to complex visual observations. This transition introduces partial observability and demands robust world modeling. We ask: Can VLM agents construct internal world models through explicit visual state reasoning? To address this question, we architecturally enforce and reward the agent's reasoning process via reinforcement learning (RL), formulating it as a Partially Observable Markov Decision Process (POMDP). We find that decomposing the agent's reasoning into State Estimation ("what is the current state?") and Transition Modeling ("what comes next?") is critical for success, as demonstrated through five reasoning strategies. Our investigation into how agents represent internal beliefs reveals that the optimal representation is task-dependent: Natural Language excels at capturing semantic relationships in general tasks, while Structured formats are indispensable for precise manipulation and control. Building on these insights, we design a World Modeling Reward that provides dense, turn-level supervision for accurate state prediction, and introduce Bi-Level General Advantage Estimation (Bi-Level GAE) for turn-aware credit assignment. Through this form of visual state reasoning, a 3B-parameter model achieves a score of 0.82 across five diverse agent benchmarks, representing a 3$\times$ improvement over its untrained counterpart (0.21) and outperforming proprietary reasoning models such as GPT-5 (0.75), Gemini 2.5 Pro (0.67) and Claude 4.5 (0.62). All experiments are conducted within our VAGEN framework, a scalable system for training and analyzing multi-turn VLM agents in diverse visual environments. Code and data are publicly available at https://vagen-ai.github.io.

Related papers

• The Hierarchy of Agentic Capabilities: Evaluating Frontier Models on Realistic RL Environments [0.11586753333439907]
  We present an empirical study evaluating frontier AI models on 150 workplace tasks within a realistic e-commerce RL environment from Surge.<n>Our analysis reveals an empirically-derived emphhierarchy of agentic capabilities that models must master for real-world deployment.<n>Weaker models struggle with fundamental tool use and planning, whereas stronger models primarily fail on tasks requiring contextual inference beyond explicit instructions.
  arXiv  Detail & Related papers  (2026-01-13T23:49:06Z)
• Agent2World: Learning to Generate Symbolic World Models via Adaptive Multi-Agent Feedback [51.22403664895878]
  Agent2World is a tool-augmented multi-agent framework that achieves strong inference-time world-model generation.<n>It also serves as a data engine for supervised fine-tuning, by grounding generation in multi-agent feedback.
  arXiv  Detail & Related papers  (2025-12-26T18:54:14Z)
• Do Vision-Language Models Have Internal World Models? Towards an Atomic Evaluation [54.3628937181904]
  Internal world models (WMs) enable agents to understand the world's state and predict transitions.<n>Recent large Vision-Language Models (VLMs), such as OpenAI o3, GPT-4o and Gemini, exhibit potential as general-purpose WMs.
  arXiv  Detail & Related papers  (2025-06-27T03:24:29Z)
• GAM-Agent: Game-Theoretic and Uncertainty-Aware Collaboration for Complex Visual Reasoning [9.24980723036746]
  GAM-Agent is a game-theoretic multi-agent framework for enhancing vision-language reasoning.<n>It formulates the reasoning process as a non-zero-sum game between base agents--each specializing in visual perception subtasks--and a critical agent that verifies logic consistency and factual correctness.<n>Our approach is modular, scalable, and generalizable, offering a path toward reliable and explainable multi-agent multimodal reasoning.
  arXiv  Detail & Related papers  (2025-05-29T12:37:34Z)
• Boosting Virtual Agent Learning and Reasoning: A Step-Wise, Multi-Dimensional, and Generalist Reward Model with Benchmark [72.46357004059661]
  Generalist Virtual Agents (GVAs) have shown significant promise in autonomous task execution.<n>To address these challenges, we propose Similar, a Step-Wise Multi-Dimensional Generalist Reward Model.<n>Similar offers fine-grained signals for agent training and can choose better action for inference-time scaling.
  arXiv  Detail & Related papers  (2025-03-24T13:30:47Z)
• VisualPRM: An Effective Process Reward Model for Multimodal Reasoning [76.35753243272521]
  We introduce VisualPRM, which improves the reasoning abilities of existing Multimodal Large Language Models (MLLMs)<n>Our model achieves a 5.9-point improvement across seven multimodal reasoning benchmarks.<n>For the evaluation of multimodal PRMs, we propose VisualProcessBench, a benchmark with human-annotated step-wise correctness labels.
  arXiv  Detail & Related papers  (2025-03-13T12:03:37Z)
• Expanding Performance Boundaries of Open-Source Multimodal Models with Model, Data, and Test-Time Scaling [191.7830199016589]
  We introduce InternVL 2.5, an advanced multimodal large language model (MLLM) series that builds upon InternVL 2.0.<n>InternVL 2.5 exhibits competitive performance, rivaling leading commercial models such as GPT-4o and Claude-3.5-Sonnet.<n>We hope this model contributes to the open-source community by setting new standards for developing and applying multimodal AI systems.
  arXiv  Detail & Related papers  (2024-12-06T18:57:08Z)
• Large Language Model Evaluation Via Multi AI Agents: Preliminary results [3.8066447473175304]
  We introduce a novel multi-agent AI model that aims to assess and compare the performance of various Large Language Models (LLMs) Our model consists of eight distinct AI agents, each responsible for retrieving code based on a common description from different advanced language models. We integrate the HumanEval benchmark into our verification agent to assess the generated code's performance, providing insights into their respective capabilities and efficiencies.
  arXiv  Detail & Related papers  (2024-04-01T10:06:04Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.