Digest

Reset

1180 papers

2026-09-30 · arXiv:2609.33566

Correct then Forecast: Observer State-Space Models for Time Series Forecasting

Alexis-Raja Brachet, Guillaume Clavier--Fr\'emond, Abdelhakim Ziani et al.

The authors introduce Observer State-Space Models (OSSMs) for time-series forecasting. OSSMs treat the observed input sequence as measurements correcting an internal latent state via an observer, while the latent transition is autonomous and shared across context and forecasting intervals. They claim OSSM improves forecasting versus corresponding SSMs under the same parameter count and training setup.

Latent DynamicsAI summary
2026-09-30 · arXiv:2609.33595

Beyond One-Step Accuracy: State-Affine Latent Transition for Reliable Visual Planning

Boyuan Zhang, Yingjun Du, Xiantong Zhen et al.

The authors argue that one-step prediction loss does not reflect how errors propagate during recursive latent rollouts used for visual planning. They analyze when rollout error propagation becomes independent of the rollout state and introduce SALT, a state-affine latent transition trained with recursive multi-step rollout supervision. Across four visual planning environments, SALT improves closed-loop success while having higher one-step prediction error than LeWM.

Latent DynamicsCode availableAI summary
2026-09-30 · arXiv:2609.33728

ALDER: Discovering the Laws of a World by Acting in It

Teng Cao, Yu Deng, Quentin Delfosse et al.

ALDER discovers and revises explicit, equation-based world models by acting in the environment. It proposes parametric equation structures, fits coefficients, and uses an independent verifier on held-out data. When multiple hypotheses fit, an experiment selector queries cost- and safety-aware interventions where predictions disagree. Counterexamples update evidence and drive structural revisions; validated equations are used for prediction and control via an inverse problem.

DrivingAI summary
2026-09-30 · arXiv:2609.33804

MinkowskiPE: Minkowski Positional Encoding for Spatiotemporal Perception

Yuhao Li, Louie Hong Yao, Tianyi Shi et al.

MinkowskiPE proposes Minkowski positional encoding for spatiotemporal perception. It assigns each token a spacetime coordinate and uses the relative spacetime displacement between tokens to parameterize Lorentz transformations applied to query/key features, making attention depend on position only through relative displacement and be translation-invariant. It retains the standard dot-product attention interface and is compatible with efficient attention implementations.

2026-09-30 · arXiv:2609.33832

Achieve What You Imagined: Learning to Align Actions with Visual Plans

Yuheng Qiao, Ziran Wei, Xiaohan Wang et al.

The authors treat a world-action model’s visual predictions as a goal-conditioned visual proposal and use a frozen action-conditioned world model to predict action consequences. They compute dense feedback from cross-model prediction consistency plus terminal goal alignment, then use Flow Policy Optimization to update only the action head. Across four real-world UR5 tasks, they report mean success increasing from 43.4% to 75.1%, without online robot interaction or task-specific reward models.

DrivingAI summary
2026-09-30 · arXiv:2609.33844

ViBR-WM: Visual Bayesian Regression for World Modeling

Jifan Li, Ning Ning

ViBR-WM is a visual world model that forecasts in a DINOv2-based latent representation and uses Bayesian regression to predict joint visual–physical states recursively and physical targets directly. It combines trend, seasonal, and cycle modules with interpretable regression and Bayesian variable selection, averaging over predictor subsets via posterior probabilities and modeling parameter uncertainty and future disturbances.

Latent DynamicsAI summary
2026-09-30 · arXiv:2609.33848

QwenGyre: An Elastic Reinforcement Learning Framework for Training xLong-Horizon Agents

Weiqi Wang, Yuxin Zhou, Mouxiang Chen et al.

QwenGyre is an end-to-end framework for x Long-horizon online RL with black-box LLM agent harnesses. It uses an elastic scheduler to elastically reallocate GPUs between rollout and training without interrupting live executions, and a trajectory processor that reconstructs branching histories, scores partial progress, and deduplicates redundant paths to bound training cost. On NL2RepoBench with Qwen3.8 2.4T, it improves score in 48 steps and reports end-to-end speedups over Colocate and Async.

DrivingAI summary
2026-09-30 · arXiv:2609.33940

Behavioral Monitoring of JEPA World Models with Jacobian Centroids

Thomas Walker, Randall Balestriero, Richard Baraniuk

The authors propose behavioral monitoring signals for JEPA world models using centroids defined as sub-component Jacobian row-sums. Centroids can be computed via Jacobian vector products, characterize internal input-space geometry, and yield saliency maps. On Push-T and TwoRoom under distribution shift, a structural dissociation (encoder goal represented, predictor unresponsive) predicts planning failure before actions; centroid-based methods outperform activation- and reconstruction-based shift detectors, enabling pre-execution goal resampling.

JEPAAI summary
2026-09-30 · arXiv:2609.34058

Do World Models Learn Global Understanding?

Alexander Detkov, Matt Thomson

The authors study whether models learn global constraints and propagate their consequences from local transition data. In controlled monoid worlds, next-state training fits paths but fails to propagate non-trivial constraints, while compositional training (hiding intermediates) achieves 96% average accuracy on inverse/commutativity/composition. Generalization declines sharply with proof depth; longer compositional path length improves deeper propagation and improves related results in vision and language.

Latent DynamicsAI summary
2026-09-30 · arXiv:2609.34085

AD-E2E-JEPA: A Joint-Embedding Predictive Architecture For End-to-End Autonomous Driving

Haoran Zhu, Wancong Zhang, Yann LeCun et al.

AD-E2E-JEPA is an action-conditioned JEPA world model for end-to-end autonomous driving that enables goal-conditioned zero-shot planning without training a driving policy. It adds a SIGReg-regularized learnable projector that compresses patch embeddings, reducing planning latency (0.8s for an 8-frame rollout over 256 trajectories) while retaining planning performance on NAVSIMv2.

DrivingCode availableAI summary
2026-09-30 · arXiv:2609.34159

WorldGraph: Graph-Native World Modeling

Zezhong Ding, Yipeng Li, Xike Xie

WorldGraph studies graph world modeling (GWM) where the evolving graph itself is the world dynamics. It formulates latent graph states and heterogeneous transition prediction over node-, edge-, and graph-level changes, builds the GWM-Zero benchmark (8 temporal graph datasets), and proposes WorldGraph with a state-aware graph transformer and a transition-aware GRPO (dynamic grouping, structure-aware verifiable rewards).

JEPAAI summary
2026-09-30 · arXiv:2609.34286

Dexterous Tactile World Model

Ziyao Zeng, Xiatao Sun, Hao Wang et al.

DTWM is a video world model for future-frame prediction of egocentric human manipulation using both RGB video and dexterous tactile glove signals from each hand. It conditions a pretrained video diffusion transformer by injecting a zero-initialized tactile residual at each hand’s location in the video tokens with block-causal masking so predicted frames cannot use future info. Compared to a matched vision-only baseline, it reduces hand-motion underestimation (23% to 9%) and hand-region perceptual error.

2026-09-30 · arXiv:2609.34375

LRC-JEPA: Disentangling Dynamics and Residual Context for Efficient World Models

Luzhe Huang, Lei Chu, Jingyi Liang et al.

LRC-JEPA is a compact JEPA world model that routes information into two streams: a predictive latent z for action-conditioned dynamics and planning, and a residual context embedding u for reconstruction. Only z is rolled out by the dynamics model at test time. The authors report improved planning success in simulated control and better Bridge-v2 offline action-recovery while using a 5.5M active-parameter encoder and faster planning.

JEPAAI summary
2026-09-30 · arXiv:2609.34391

P2P: Cross-View Population Denoising for Unpaired Single-Cell Perturbation Response Prediction

Haojie Yang, Ran Su

P2P predicts unpaired single-cell perturbation response at the population level by learning shared condition effects from two stochastic “views” of the same control/perturbed condition. It encodes control sets with a permutation-invariant set encoder, perturbations with a structured encoder, and blends an empirical memory with a neural residual to output population mean and gene-wise variance. Trained with cross-view supervision (no cell correspondences).

DrivingAI summary
2026-09-30 · arXiv:2609.34604

Shaping Persistent Representations from Independent Interactions

Ji Dai, Quan Fang, Junyu Gao et al.

SPRII (Shaping Persistent Representations from Independent Interactions) is a relation-supervised training principle for world models that learns persistent context from related independent interactions without numerical property labels. It uses two components: Align (related contexts agree) and Cross (use one interaction’s context to predict another’s future), while keeping the learner’s native objective. The paper analyzes Formation, Use, and Value links from accessible persistent information to task error reduction.

Latent DynamicsAI summary
2026-09-30 · arXiv:2609.34677

Learning What to Recall: Adaptive Multi-Cue Episodic Memory for World Models

Beomsu Kim, Chieh-Hsin Lai, Bac Nguyen et al.

The authors propose Future-Aware Recall (FAR), a framework for episodic memory access in world models. FAR learns which past memories to retrieve and which retrieval cues (time, pose, vision, audio) to trust by supervising a future-blind retriever using future-aware predictive utility during training (approximated by negative diffusion prediction loss).

2026-09-30 · arXiv:2609.34911

Don't Throw Away the Tail: Action Upcycling for Policy Acceleration

Taesung Kwon, Jangho Park, Sunwoo Park et al.

Action Upcycling is a training-free way to extend the execution horizon of chunked robot policies by reusing “tail” actions the policy would otherwise discard. Using a velocity-fluctuation signal computed from a single sampled action chunk (no model internals, no extra samples), it executes tail actions until the predicted motion’s velocity starts to fluctuate. It reduces policy calls by 1.2–1.7× with no loss in success rate across several VLA models and a World Action Model, and works on real robots.

DrivingAI summary
2026-09-30 · arXiv:2609.35047

EMPIRIC: Experiment-Driven Learning of Residual World Models for Robot Planning

Yichao Liang, Amber Li, Dat Nguyen et al.

EMPIRIC learns residual world models for robot planning by extending a base physics simulator with Python programs for missing physical mechanisms (forces, constraints, recurrent hidden state). It uses Bayesian inference to estimate program parameters and hidden state from noisy observations, chooses informative experiments, revises the programs when predictions fail, and plans under uncertainty. It reports solving more tasks with fewer environment interactions than three baselines across five simulated domains, and demonstrates on a physical robot.

DrivingAI summary
2026-09-30 · arXiv:2609.35138

FlexiWorld: Learning and Planning via Flexible Action Chunks Across Multiple Time Scales

Shidu Ren, Qilin Gu, Zhenghao Ni et al.

FlexiWorld is a JEPA-based latent world model for goal-directed planning that learns from mixed-span goal supervision while using variable-length action chunks across multiple time scales. It jointly trains a flexible (variable-length) causal action encoder, latent predictor, and an autoregressive goal-conditioned actor with Student Forcing. For planning, it proposes ARCEM (actor-residual CEM), combining action-residual search with within-chunk feedback and chunk-boundary latent prediction.

JEPAAI summary
2026-09-30 · arXiv:2609.35288

$\lambda$-JEPA Spectral Anti-Collapse Regularization for Self-Supervised Learning

Berker Demirel, Cl\'ementine Domin\'e, Valentino Maiorca et al.

The authors introduce SACReg, a spectral anti-collapse regularizer meant to prevent dimensional collapse in the backbone representation (not just the projected space) by using a log-determinant penalty on representation covariance (and a mean penalty). They derive the idea from a λ-balance analysis in a two-layer linear network, extend it to nonlinear encoders, and apply it to JEPA as λ-JEPA with SACReg on both backbone and projector. They report improved classification/transfer and video SSL results, with released code.

JEPACode availableAI summary