Latest World Models Research Papers

The newest World Models papers from across the field — arXiv, NeurIPS, CVPR, Nature, and more — refreshed daily and ranked by relevance. Distill AI tracks World Models so you don’t have to: get the standout work delivered to your inbox every morning, with 2-sentence summaries and the option to chat with any paper.

Recent papers

• world2vec: Mechanistic world models through latent dynamics learning
  Tobias Schmidt, Rodrigo Gonzalez Laiz, Krisztián Szuppinger, Steffen Schneider · arXiv · Oct 29, 2026

  World models can achieve strong control performance without faithfully capturing the dynamics of their environment, limiting their potential for mechanistic interpretability and scientific discovery. Here, we introduce world2vec to learn me…

• Homebrewed SLMs: Autoregressive and Masked Diffusion Language Models Built at Home for Portuguese
  Pedro Carvalho Brom · arXiv · Oct 29, 2026

  Building a language model of one's own no longer requires a data centre. This tutorial trains one from scratch, tokenizer included, on one consumer GPU, and shows how to tell whether it is good for the tasks of a specialist domain. It propo…

• Rural Readiness: Domain Generalization with Adversarially Chosen Domains
  Cynthia Dwork, Andisheh Ghasemi, Lunjia Hu · arXiv · Oct 29, 2026

  Predictive models trained at well-resourced data-rich institutions often fail when deployed in new populations. Motivated by the goal of building clinical diagnosis and decision-making algorithms for data-scarce rural populations using the …

• Understanding the limits of latent world models
  Rodrigo Gonzalez Laiz, Steffen Schneider · arXiv · Oct 27, 2026

  World models are widely proposed as core components for enabling agents to reason over long temporal and spatial horizons. Yet current evaluations often emphasize task performance while providing limited insight into whether learned world m…

• PDRVLA: A Vision-Language-Action Model with Physical Dynamics Representations
  Hanqing Wang, Zhaotian Cai, Ruichuan An, Bohan Zeng et al. · arXiv · Oct 27, 2026

  Vision--Language--Action (VLA) models generate robot actions from visual observations and language instructions, but their understanding of physical dynamics remains limited. Video world models can capture scene dynamics, yet generating fut…

• Kronecker Delta Rule for Long Context Linear Attention
  Anamaria-Roberta Hartl, Levente Zólyomi, Michael List, Günter Klambauer et al. · arXiv · Oct 19, 2026

  We propose KronLSTM, a linear attention architecture based on the Kronecker Delta Rule, a novel update rule for associative matrix memories. KronLSTM extends the xLSTM family with a memory-revision mechanism that corrects targeted associati…

• Agent World Model in the Wild: Computer-Use Agent Environment Synthesis
  Zhaoyang Wang, Qianhui Wu, Haorui Wang, Simon Yu et al. · arXiv · Oct 19, 2026

  Agent learning is shifting from learning from human data toward learning through interaction with the world. However, scaling such learning directly in real-world computer systems is challenging and can introduce substantial security risks.…

• Simulation Priors for Rapid Adaptation in Model-Based RL for Autonomous Racing
  Adrian Remonda, Jiajun Xi, Nicklas Hansen, Marcus Greiff et al. · arXiv · Oct 19, 2026

  Reinforcement learning enables learning high-performance control policies, but its reliance on extensive interaction limits deployment on physical systems where data collection is costly, and failures are unsafe. We introduce Prior-Informed…

• GeoState: Learning View-Decoupled Geometry States for 3D World Modeling
  Hongchang Liu, Jian Liu, Yujuan Ding, Liang Peng et al. · arXiv · Oct 15, 2026

  Recent 3D world models increasingly use features from geometry foundation models (GFMs) as predictive world states, benefiting from their rich geometric priors. However, the view dependence of these features entangles scene dynamics with ca…

• Revisiting Simple Baselines for Relational Deep Learning
  Viet The Nguyen, Vijay Prakash Dwivedi, Gunther Gust · arXiv · Oct 14, 2026

  Relational deep learning has emerged as a powerful paradigm for learning end-to-end predictive models over relational databases by representing them as temporal heterogeneous graphs. Recent advances based on graph neural networks and Transf…

• Learning When to Transfer: Reinforcement Learning for Carbon-Footprint Modeling in New Manufacturing Domains
  Muhammad Umar Farooq, Daniel R. Cooper · NeurIPS 2026 · Oct 13, 2026

  Manufacturers often need to characterize a new material, process, or product before enough operating data are available to train reliable predictive models for this new, target domain. Existing manufacturing operations contain potentially u…

• UniW(A)M: A Unified Mixture-of-Experts Framework for World and World-Action Models
  Jingbo Xu, Jianjie Fang, Zhuohang Li, Zihao Song et al. · arXiv · Oct 13, 2026

  World models (WMs) and world-action models (WAMs) focus on future world prediction and executable action generation, respectively. Although these models learn shared physical dynamics and interaction semantics from embodied trajectories, th…

• Dream to Remember: Continual Steering of Generalist Robot Policies with World Models
  Bahey Tharwat, Iman Nematollahi, Marvin Alles, Xingyuan Zhang et al. · arXiv · Oct 9, 2026

  Generalist robot policies are trained with imitation learning and are bounded by the quality and coverage of their demonstrations. Reinforcement learning can improve and extend them, either by fine-tuning the policy weights or by training a…

• WOVEN: Weaving Visual World Modeling into Multimodal LLMs
  Zheyu Fan, Yue Zhang, Mingkai Deng, Kangrui Wang et al. · arXiv · Oct 8, 2026

  Multimodal large language models (MLLMs) struggle with spatial, embodied, physical, and temporal reasoning. We hypothesize that these failures reflect a shared deficit in visual transition reasoning, and test whether this capability can ser…

• RiCo: Neural Simulation of Rigid-Body Interactions via Local Contact Reasoning
  Ruixiang Ouyang, Guanren Qiao, Fansen Meng, Yueci Deng et al. · arXiv · Oct 8, 2026

  Accurate simulation of rigid-body interactions is essential for predictive physical world models. Despite recent progress in modeling object dynamics, capturing how local contacts between surfaces shape object motion remains challenging. Wh…

• Decoupling Exploration from Optimization in RLVR
  Saif Punjwani, Micah Goldblum · arXiv · Oct 7, 2026

  Modern language models undergo reinforcement learning with verifiable rewards (RLVR) on top of already-trained checkpoints. A key promise of RLVR is the discovery of new reasoning strategies. In principle, a model can sample novel ideas abs…

• AdvSim2Real : Training Web Agents Against Adaptive Prompt Injection in a Web World Model
  Sarim Hashmi, Mukul Ranjan, Kshitij Mishra, Mikhail Kuznetsov et al. · arXiv · Oct 6, 2026

  Web agents complete user requests by reading and acting on pages that third parties write, so an instruction planted on a page can redirect the agent away from the user's goal. The agent cannot simply ignore the page, because the page also …

• H-JEPA: End-to-End Learning of Hierarchical World Models for Visual Planning
  Wancong Zhang, Basile Terver, Michael Rabbat, Yann LeCun et al. · arXiv · Oct 5, 2026

  Long-horizon planning with latent world models requires reasoning across timescales and levels of abstraction. Existing task-agnostic JEPA world models predict and plan at a single timescale or with multiple horizons in one shared latent sp…

• IdeaLens: Detecting AI Ideas in Long-form Writing
  Rishanth Rajendhran, Minjoon Choi, Jenna Russell, Ramya Namuduri et al. · arXiv · Oct 5, 2026

  While modern AI detectors identify who wrote the words, emerging policies on AI use increasingly hinge on a different question: who came up with the ideas? We introduce IdeaLens, a detector that identifies whether a document's ideas came fr…

• What Should World Models Forget? Stratified Retention for Continual Adaptation
  Nishit Anand, Ramani Duraiswami, Dinesh Manocha · arXiv · Oct 2, 2026

  Continual learning treats degradation on previously seen data as evidence of failure, a convention inherited from settings with a stationary prediction target, where a correct label remains correct indefinitely. World models do not satisfy …

• ReCIRC: Rectified Conformal Risk Control
  Bruno Marcondes e Resende, Helton Graziadei, Thiago Rodrigo Ramos, Rafael Izbicki · arXiv · Sep 29, 2026

  Many applications of black-box predictive models require controlling task-relevant error rates, such as missed lesion pixels in segmentation or missed labels in multilabel classification. Conformal risk control (CRC; Angelopoulos et al., ar…

• Control-Geometry Straightening for Sampling-Based Latent Planning
  Ziang Fu, Ning Ning · arXiv · Sep 28, 2026

  Joint-embedding predictive architectures enable planning with latent world models, but accurate transition prediction alone does not ensure that the planning objective is easy to optimize. We introduce Control-Geometry Straightening (CGS), …

• Retrainable physics-integrated neural differentiable modeling of sintering across material systems
  Zeping Chen, Ani Aprahamian, Khachatur V. Manukyan, Tengfei Luo · arXiv · Sep 25, 2026

  Sintering is widely used to manufacture ceramics, but coupled densification and grain growth, material-dependent kinetics, and sparse measurements complicate predictive modeling and process design. We present Sinter-PiNDiff, a retrainable p…

• Aim Short to Reach Far: Your Frozen World Model Can Plan Better Than You Think
  Xvyuan Liu, Jianjie Fang, Chen Gao, Yong Li · arXiv · Sep 24, 2026

  Planners built on visual world models commonly score each predicted outcome by its distance to the encoded goal image. We show that this target can limit control even with exact dynamics and globally optimal short-horizon search: reaching a…

• Agent-Editing World Model: Rethinking World Modeling for LLM Agents
  Shuang Sun, Guoxin Chen, Fanzhe Meng, Jia Deng et al. · arXiv · Sep 23, 2026

  Recent advances in large language models (LLMs) have enabled agents to tackle long-horizon tasks across diverse environments. To further improve agent performance, existing language world models typically predict environment observations, y…

• Generalizable Robotic Insertion with World Models
  Nicklas Hansen, Iretiayo Akinola, Yijie Guo, Jie Xu et al. · arXiv · Sep 23, 2026

  Robotic assembly in high-mixture settings requires adaptable systems that can handle diverse parts, yet current approaches typically rely on policies specialized to each insertion task. Although this can reach high success rates, it makes t…

• hyperbolix: Hyperbolic Deep Learning in JAX
  Timo Klein, Thomas Lang, Yllka Velaj, Sebastian Tschiatschek · arXiv · Sep 23, 2026

  We present hyperbolix, an open-source library for hyperbolic deep learning in JAX, built on Flax NNX. To our knowledge, it is the first comprehensive, general-purpose hyperbolic deep learning library in JAX. It includes six manifolds with a…

• D-JEPA: A Decision-Aligned Latent World Model
  Shuaijun Liu, Chengyu Wu, Qifu Wen, Feiyang You et al. · arXiv · Sep 21, 2026

  Latent world models predict the consequences of actions, but accurate prediction does not guarantee that latent distance reflects which candidate will execute successfully. We identify a decision-local prediction gap: among the few futures …

• API World Model: Agentic, Programmable, and Interactive World Model with Agent-Driven Virtual-World Rollouts
  ICLR 2027 Conference Submission · Sep 19, 2026

  World models are just at a tipping point where they could really take off, with the promise of enabling agents to understand, predict, and interact with the physical world. Recent efforts have largely pursued this vision by leveraging video…

• Benchmarking World Models for Continual Learning on Compositional Tasks
  Haoyu Zhou, Joe Watson, Anson Lei, Ingmar Posner · arXiv · Sep 18, 2026

  A desirable property of a world model is the ability to learn continually across tasks, adapting to new environments without forgetting what the agent has already learnt. In particular, the ability to retain and reuse knowledge obtained fro…