~ similar to 2607.27201· 19 results
This paper introduces the Mental World Model and its implementation, the Multi-Branch Mental Evolution Architecture, to simulate and model subjective cognition and mental states in artificial intellig…
Chaehyun Kim, Chi-Pin Huang, Jisoo Kim, Min-Hung Chen +3 more
This paper introduces ThinkWorld, a VLA learning framework that grounds action reasoning in world dynamics and strengthens it through reinforcement learning with world imagination.
Adam J. Thorpe, Stepan Tretiakov, Cheng-Hsi Hsiao, Su Ann Low +5 more
The paper argues that for embodied AI to be safe and effective, world models must be physically viable, requiring a structural shift from mere observation prediction to representing the underlying phy…
Lanxiang Hu, Yixin Huang, Abhilash Shankarampeta, Zilin Dai +12 more
This paper introduces WMEval, a benchmark for evaluating interactive world models against physical simulations.
Jinpeng Lu, Dexu Zhu, Haoyuan Shi, Linghan Cai +7 more
WRBench introduces a new benchmark for evaluating world models
Yuke Zhao, Wangbo Zhao, Weijie Wang, Zeyu Zhang +7 more
WorldOlympiad is introduced as a benchmark for evaluating video-based world models across physical faithfulness, geometric consistency, and interaction fidelity.
Jiabin Qiu, Zixuan Chen, Hongye Cao, Jieqi Shi +2 more
This paper introduces AD-WM, a world model that preserves action-dependent differences for counterfactual selection in model predictive control.
The paper introduces NEST, a graph-theoretic representational ontology for modeling cognition as structured state formation and transformation.
Zhikai Pan, Chih-Ting Liao, Chunrui Liu, Xi Xiao +4 more
The paper introduces a multilingual benchmark (MentalMap) to test if LLMs build internal spatial world models from text, finding a universal 'L3 reasoning cliff' suggesting that text-only working memo…
The paper argues that large language models (LLMs) are a special case of world models and proposes a continuous spectrum between token prediction and latent-space architectures.
Ruoxuan Zhang, Qiaoqiao Wan, Zhengguang Wang, Chenghao Yu +3 more
The paper introduces MindClaw, a closed-loop framework that enables embodied agents to perform real-time mental-state reasoning and intervene with precision, significantly outperforming standard VLM b…
The paper proposes a Multi-Phase Inference Mechanism (MIM) to formalize how diverse world models arise, reframing alignment as making heterogeneous representations mutually processable rather than for…
This paper proposes robot-factored world models for action-conditioned video prediction in robotics, which factor out action realization and robot rendering to avoid learning the realization process a…
Chenming Zhu, Jingli Lin, Yilin Long, Peizhou Cao +3 more
The paper proposes Astra, an agentic framework that equips Vision-Language Models (VLMs) with the ability to perform spatial reasoning by actively generating and utilizing imagined visual evidence fro…
Kevin Wang, Anna Thöni, Benjamin Kempinski, Bobby Cheng +49 more
The paper introduces Mindgames, a comprehensive multi-game arena for evaluating LLM agents' sustained social and strategic reasoning, demonstrating that current evaluations are limited by structural s…
Junwon Seo, Sushant Veer, Ran Tian, Wenhao Ding +5 more
StressDream proposes a novel method to steer video world model imaginations toward high-impact, yet plausible outcomes, enabling robust policy evaluation and improvement by identifying undesirable fut…
The paper proposes extending world models for multi-agent reinforcement learning by factorizing the latent state to explicitly model and predict the unobservable intentions and behaviors of teammates.
Tianzhuo Yang, Zihan Shen, Zirui Mi, Zhaoyi Zhang +6 more
The paper introduces MiraBench, a new benchmark that evaluates the action-conditioned reliability of robotic world models, finding that visual fidelity is insufficient and that optimism bias is a perv…