# Semantic Belief-State World Model for 3D Human Motion Prediction

arXiv: https://arxiv.org/abs/2601.03517 · HTML: https://mlsanity.com/paper/2601.03517/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Placed in Embodied and robot world models, one of 11 directions in World models.
- Builds on 25 of its 34 references that are in the map.
- First posted on arXiv 7 Jan 2026.

## Placed in

- World models / [Embodied and robot world models](https://mlsanity.com/wm/embodied-and-robot-world-models/)

## Builds on (most cited first, 25 of 25)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [Dream to Control: Learning Behaviors by Latent Imagination](https://mlsanity.com/paper/1912.01603/) (2019)
- [World Models](https://mlsanity.com/paper/1803.10122/) (2018)
- [Mastering Atari, Go, Chess and Shogi by Planning with a Learned Model](https://mlsanity.com/paper/1911.08265/) (2019)
- [Learning Latent Dynamics for Planning from Pixels](https://mlsanity.com/paper/1811.04551/) (2018)
- [Mastering Diverse Domains through World Models](https://mlsanity.com/paper/2301.04104/) (2023)
- [Spatial Temporal Graph Convolutional Networks for Skeleton-Based Action Recognition](https://mlsanity.com/paper/1801.07455/) (2018)
- [Human Motion Diffusion Model](https://mlsanity.com/paper/2209.14916/) (2022)
- [Mastering Atari with Discrete World Models](https://mlsanity.com/paper/2010.02193/) (2020)
- [Scheduled Sampling for Sequence Prediction with Recurrent Neural Networks](https://mlsanity.com/paper/1506.03099/) (2015)
- [MotionDiffuse: Text-Driven Human Motion Generation with Diffusion Model](https://mlsanity.com/paper/2208.15001/) (2022)
- [VideoGPT: Video Generation using VQ-VAE and Transformers](https://mlsanity.com/paper/2104.10157/) (2021)
- [Improving Variational Inference with Inverse Autoregressive Flow](https://mlsanity.com/paper/1606.04934/) (2016)
- [T2M-GPT: Generating Human Motion from Textual Descriptions with Discrete Representations](https://mlsanity.com/paper/2301.06052/) (2023)
- [MotionGPT: Human Motion as a Foreign Language](https://mlsanity.com/paper/2306.14795/) (2023)
- [Action-Conditioned 3D Human Motion Synthesis with Transformer VAE](https://mlsanity.com/paper/2104.05670/) (2021)
- [PhysDiff: Physics-Guided Human Motion Diffusion Model](https://mlsanity.com/paper/2212.02500/) (2022)
- [On human motion prediction using recurrent neural networks](https://mlsanity.com/paper/1705.02445/) (2017)
- [HuMoR: 3D Human Motion Model for Robust Pose Estimation](https://mlsanity.com/paper/2105.04668/) (2021)
- [ReMoDiffuse: Retrieval-Augmented Motion Diffusion Model](https://mlsanity.com/paper/2304.01116/) (2023)
- [Learning Trajectory Dependencies for Human Motion Prediction](https://mlsanity.com/paper/1908.05436/) (2019)
- [MoFusion: A Framework for Denoising-Diffusion-based Motion Synthesis](https://mlsanity.com/paper/2212.04495/) (2022)
- [Perpetual Humanoid Control for Real-time Simulated Avatars](https://mlsanity.com/paper/2305.06456/) (2023)
- [QuaterNet: A Quaternion-based Recurrent Model for Human Motion](https://mlsanity.com/paper/1805.06485/) (2018)
- [SoMoFormer: Social-Aware Motion Transformer for Multi-Person Motion Prediction](https://mlsanity.com/paper/2208.09224/) (2022)

## Built on by (most cited first, 1 of 1)

- [Recursive Belief Vision Language Action Models](https://mlsanity.com/paper/2602.20659/) (2026)

## Abstract (arXiv)

Human motion prediction has traditionally been framed as a sequence regression problem where models extrapolate future joint coordinates from observed pose histories. While effective over short horizons this approach does not separate observation reconstruction with dynamics modeling and offers no explicit representation of the latent causes governing motion. As a result, existing methods exhibit compounding drift, mean-pose collapse, and poorly calibrated uncertainty when rolled forward beyond the training regime. Here we propose a Semantic Belief-State World Model (SBWM) that reframes human motion prediction as latent dynamical simulation on the human body manifold. Rather than predicting poses directly, SBWM maintains a recurrent probabilistic belief state whose evolution is learned independently of pose reconstruction and explicitly aligned with the SMPL-X anatomical parameterization. This alignment imposes a structural information bottleneck that prevents the latent state from encoding static geometry or sensor noise, forcing it to capture motion dynamics, intent, and control-relevant structure. Inspired by belief-state world models developed for model-based reinforcement learning, SBWM adapts stochastic latent transitions and rollout-centric training to the domain of human motion. In contrast to RSSM-based, transformer, and diffusion approaches optimized for reconstruction fidelity, SBWM prioritizes stable forward simulation. We demonstrate coherent long-horizon rollouts, and competitive accuracy at substantially lower computational cost. These results suggest that treating the human body as part of the world models state space rather than its output fundamentally changes how motion is simulated, and predicted.