Video World Models Need Predictable Latents
arXiv:2610.34663v1 Announce Type: cross Abstract: Video world models trained on large-scale robot data promise a safe and scalable approach for learning and evaluating policies that act in the physic...
Contrastive World Models propose a new method for learning latent dynamics without pixel reconstruction. By replacing observation reconstruction with a Deep InfoMax-like objective that maximizes mutual information between state-action sequences and local patch features of future observations, the approach encourages state representations to retain predictive information while ignoring visually irrelevant details. Experiments show that this method matches existing baselines in simple settings and significantly outperforms them when distractors or natural video backgrounds are present, while also training more efficiently by eliminating the pixel decoder.
arXiv:2610.34663v1 Announce Type: cross Abstract: Video world models trained on large-scale robot data promise a safe and scalable approach for learning and evaluating policies that act in the physic...
StarWM introduces a self‑supervised attention routing mechanism that selectively applies reconstruction only to dynamically relevant regions of visual input. By combining a cross‑attention module with a dual‑stream decoder and stop‑gradient barriers, it balances faithful environmental dynamics capture with abstraction of irrelevant content. Experiments on DeepMind Control show that StarWM outperforms both reconstruction‑based and reconstruction‑free baselines, especially under distractor conditions, and preserves state attributes over long‑horizon imagination.
arXiv:2610.14228v1 Announce Type: cross Abstract: Pixel-space world models have recently emerged as a promising approach for GUI agents, predicting future screens prior to execution and subsequently...
MAC‑v2 is a new framework for long‑horizon model‑based value expansion from pixel observations in offline reinforcement learning. It trains a reconstruction‑free latent world model that learns representations through prediction and control objectives, avoiding the need to encode control‑irrelevant visual details. By using action chunking, MAC‑v2 reduces recursive model predictions, mitigating error accumulation and achieving strong performance on visual benchmarks such as OGBench, MMBench2, Robomimic, and LIBERO‑10.
The paper introduces motion-guided contrastive regularization to improve JEPA-style world models by preventing static collapse, where encoders overemphasize episode-specific static content. By comparing predicted latent transitions with transitions to candidate observations within the same episode and using optical-flow similarities as soft contrastive targets, the method encourages representations that preserve manipulation-relevant state changes. Experiments on OGBench-Cube and LIBERO-Object show significant gains in planning success while maintaining performance on other visual-control benchmarks.
JEPA-style world models learn action-conditioned latent dynamics without reconstructing observations, but distributional regularization alone does not ensure that their representations preserve inform...
arXiv:2608. 06544v1 Announce Type: new Abstract: World models for visual control typically learn compact latent states by reconstructing observations, implicitly encouraging representations to preserve information across the entire visual input.
The paper introduces Dwell, a model‑based reinforcement learning framework that treats a pretrained vision‑language model (VLM) as a recurrent latent transition operator. Dwell enables the VLM to predict, adjust, and act within its own latent space, allowing the agent to learn from real interactions and latent imagination without reconstructing full futures. Experiments across five interactive visual environments show that Dwell outperforms other trainable methods while significantly reducing computational demands for imagination.
arXiv:2606. 15768v1 Announce Type: cross Abstract: Vision-Language-Action models (VLAs) leverage large-scale vision-language pretraining for semantic robot control, but often lack explicit foresight into how robot actions change the scene.
arXiv:2605.07079v2 Announce Type: replace-cross Abstract: World models predict future transitions from observations and actions. Existing works predominantly focus on image generation only. Visual fe...
arXiv:2606. 31232v1 Announce Type: new Abstract: Learning visual world models for planning requires compact latent dynamics that remain sensitive to actions, yet reconstruction-free joint-embedding objectives can collapse to action-insensitive representations.
arXiv:2608.24855v1 Announce Type: new Abstract: Latent world models are inherently strong encoders that transform image pixel to latent embedding, yet existing world models still rely on online traje...