World Potential Model: Pretrained World Knowledge as Progress Potentials

Jun Zhao; Jixin Tang; Yang Shu; Jinyang Wu; Yuyang Lu; Jingqi Tong; Hao Xu; Weifeng Ge

cs.AIOct 7, 2026

World Potential Model: Pretrained World Knowledge as Progress Potentials

Authors: Jun Zhao, Jixin Tang, Yang Shu, Jinyang Wu, Yuyang Lu, Jingqi Tong, Hao Xu, Weifeng Ge, +1 more

Organizations: National University of Singapore · Fudan University · Tsinghua University · The University of Sydney

Abstract

Long-horizon language agents often receive supervision only from terminal task outcomes, leaving little signal for distinguishing productive intermediate behavior from stagnation or even regression. Rather than learning a separate value function or process reward model for every task, we ask whether pretrained models can recognize task progress from their existing world knowledge. We formalize this capability with a World Potential Model (WPM), a goal-conditioned evaluator of task-relative realized progress in agent contexts. In ALFWorld and ScienceWorld, off-the-shelf pretrained models substantially outperform chance at recovering realized-progress structure without task-specific evaluator fine-tuning. We further anchor these progress judgments to task-specific milestones to obtain scalar world potentials, whose temporal differences provide process-sensitive step-level credit for policy optimization. Under matched comparisons, WPM-guided optimization improves success over outcome-only GRPO across all evaluated configurations. Together, these results provide initial evidence that pretrained world knowledge can support reusable realized-progress evaluation and provide useful supervision for long-horizon agents.

Figures & tables

Figure 1 : Overview of the World Potential Model ( WPM ) and its downstream scalarization. A WPM elicits task-relative realized-progress structure from pretrained world knowledge, with ordinal progress judgments as its primary interface. For downstream use, these judgments can be referenced to task-specific progress anchors to obtain scalar world potentials, whose temporal differences provide process-sensitive credit for policy optimization.

Model	ALFWorld		ScienceWorld		Avg.
Model	Pairwise	Anchored	Pairwise	Anchored	Avg.
Random Guess	50.00	20.00	50.00	15.27	33.82
MiMo-VL-7B XiaomiMiMo (2025)	92.67	58.00	97.33	63.33	77.83
Qwen2.5-VL-7B Qwen Team (2025b)	80.00	60.00	81.33	48.00	67.33
Qwen2.5-32B Qwen Team (2024)	95.33	80.67	96.67	76.00	87.17
Qwen2.5-VL-72B Qwen Team (2025a)	93.33	82.00	94.67	81.33	87.83

Table 1: Accuracy (%) on two realized-progress evaluations. Pairwise comparison directly evaluates the ordinal progress interface of a WPM , while Anchored localization evaluates context localization relative to a task-specific milestone ladder. Random Guess denotes uniform selection among the valid candidate labels for each instance and is averaged over the corresponding evaluation set.

Model	Method	Alf World	Science World
Qwen2.5‑3B	Vanilla	10.93	8.59
	GRPO	53.90	51.56
	WPM	60.50	55.39
Qwen2.5‑7B	Vanilla	15.62	13.28
	GRPO	68.70	66.40
	WPM	72.10	74.73

Table 2 : Task success rate (%) for vanilla policies, outcome-only GRPO, and WPM -guided optimization in ALFWorld and ScienceWorld . Results for GRPO and WPM are averaged over three independent random seeds.

Explore similar work

Self-Evolving World Models for LLM Agent Planning
Jun 29, 2026Xuan Zhang, Wenxuan Zhang, See-Kiong Ng +1Test-Time Adaptation World Model Learning
National University of Singapore · Singapore University of Technology and Design · Singapore Management University
Qwen-AgentWorld: Language World Models for General Agents
Jun 23, 2026Yuxin Zuo, Zikai Xiao, Li Sheng +30LLM World Models Agentic RL
Qwen Team
MemWM: Memory-Augmented Text-Based World Model
Aug 7, 2026Yujun Wang, Tao Zhang, Jinhe Bi +9LLM World Models World Model Learning
LMU Munich · Munich Center for Machine Learning (MCML) · Huawei Heisenberg Research Center +4

Jun 29, 2026, cs.AIJ/K move · Enter open · S save

Self-Evolving World Models for LLM Agent Planning

Xuan Zhang, Wenxuan Zhang, See-Kiong Ng, Yang Deng

National University of Singapore · Singapore University of Technology and Design · Singapore Management University

Test-Time Adaptation World Model Learning World Models World Model-Based Planning LLM Agent Planning Memory-Augmented Language Model Agents

Open paper AskPDF arXiv

World models offer a principled way to equip long-horizon LLM agents with foresight: predictions of action consequences before execution. However, unreliable foresight can be ignored, misused, or even degrade downstream decision-making. In this paper, we introduce WorldEvolver, a self-evolving world model framework that revises its deployment-time context while keeping the downstream agent and all model parameters frozen. WorldEvolver integrates three modules: (i) Episodic Memory, which exploits real action transitions through retrieval-based simulation; (ii) Semantic Memory, which extracts persistent heuristic rules from prediction-observation mismatches; and (iii) Selective Foresight, which filters low-confidence predictions before integrating them into agent reasoning context. We evaluate WorldEvolver on ALFWorld and ScienceWorld, measuring world model prediction accuracy on Word2World and downstream agent success rate on AgentBoard. Extensive experiments show that WorldEvolver achieves the highest prediction accuracy across three backbones and leads other world model baselines on downstream agent success rate, demonstrating that test-time memory revision enhances both predictive fidelity and planning performance.

Abstract

Explore similar work

Self-Evolving World Models for LLM Agent Planning

Qwen-AgentWorld: Language World Models for General Agents

MemWM: Memory-Augmented Text-Based World Model