Enfold: Folding World-Generator Computation into Predictive Representations for Efficient Embodied Control

Wei-Li Zeng Yi-Tong Xing Fu-Long Liu Cheng-Qun Yang An-Tao Xiang Feng Tian Jing-Nan Gao Ji-Song Cai Xin Wang Xiao-Min Wu Yao Mu Yi-Chao Yan

2026 · arXiv.org · Vol abs/2607.26657 · 0 citations · 59 references

Computer Science

TL;DR

Enfold is presented, which transfers this computation into a representation predicted from the current visual context and language instruction, and recast a world generator as a source of predictive control representations: its future need not be materialized at every step if its internal structure can be enfolded into the present.

DOI Semantic Scholar

View source

Similar papers

#artificial intelligence Preprint Aug 2026

DELE-w0.5: Inferring Action from Future Latent State for Robotic Manipulation

The core design principle of DELE-w0.5 is to model how the physical world changes under robot actions, rather than how its visual appearance evolves frame by frame, which enables cheaper training and low-latency inference.

Fenghao Lei, Zhixiong Huang, Long Yang et al. · 2 citations

#machine learning Preprint Aug 2026

Making Latent Evolution Explicit: Operator-Structured Transitions for World Action Models

This work introduces the Latent Evolution Operator Network (LEON), which models latent evolution in a learned observable space through context-modulated operator-based propagation and additive forcing and establishes transition realization as a consequential architectural choice in latent WAMs.

Xiao-Xiao Lu, Yunlong Dong, Jia-Hao Shi et al. · 0 citations

#machine learning Preprint Sep 2026

Copper-Policy: Focus on the Representation for Robust Robot Manipulation

Copper-Policy is introduced, which learns a compact World representation with the policy rather than relying on a predefined target space and predicts future observation embeddings conditioned on task intention without reconstructing pixels through temporal joint-embedding prediction.

Ze Feng, Yi-Xu Feng, Ling-Yu Xiao et al. · 1 citation

Preprint Sep 2026

Rethinking Representations for World-Action Modeling

World-action models jointly learn robot policies and predict future observations, making the representation space an interface between control and prediction. We study the design of this space through controlled comparisons, finding that neither reconstruction fidelity nor pre-trained perceptual features alone ensure e...

Hao-Yi Jiang, Liu Liu, Xin-Jiang Wang et al. · 2 citations

Preprint Sep 2026

Beyond Future Prediction: Denoising as Generative Adaptation for Robot Control

Pretrained generative Diffusion Transformers (DiTs) capture rich pixel-level visual and language-conditioned structure through large-scale image and video generation training. A growing line of robot policies builds on this generative prior, but how it should be transferred to control remains unclear, and existing appr...

Zan-Yi Wang, Yu-Heng Lei, Deng-Yang Jiang et al. · 0 citations

Preprint Sep 2026

RepFlow: Reciprocal Supervision Improves Generation and Representation in Flow Models

Generative models learn visual structure through denoising, yet their internal states are entangled with both noise level and network depth, making it difficult to obtain a stable visual representation from the generator itself. We introduce RepFlow, which learns such a representation from the generator's evolving comp...

Wei-Li Zeng, Feng Tian, Sheng-Qi Liu et al. · 0 citations