Digest

Reset

1180 papers

2026-09-30 · arXiv:2609.35338

NeuronDiscover: Agent-in-Twin for Mechanistic Discovery in Neuronal Microenvironments with World Action Models

Haowei Xu, Wanyi Fu, Hongbin Han et al.

NeuronDiscover is an Agent-in-Twin framework for mechanistic discovery in neuronal microenvironments under “twin confounding” (mechanism change vs computational twin error). A shared mechanism-grounded World Action Model (WAM) drives prediction, intervention proposals, and observation design; independently adjudicated outcomes revise a scoped Mechanism–Intervention–Observation–Outcome (MIOY) graph, compiling supported relations into executable programs with discrepancy-adjusted acceptance bounds. Evaluated in simulated transport worlds and donor-disjoint current-clamp recordings.

Latent DynamicsAI summary
2026-09-30 · arXiv:2609.35473

Handwritten Text Recognition Lives in the High-Pixel Variance Subspace

Carlos Garrido-Munoz, Jorge Calvo-Zaragoza

For Handwritten Text Recognition, the authors report discriminative signal concentrated in high-variance pixel directions. They test six self-supervised learning (SSL) methods across six handwriting benchmarks and find pixel-grounded masked image modeling (MAE, SimMIM) achieves the lowest CER in frozen and fine-tuned settings, and benefits from real-handwriting pretraining; JEPA-style and contrastive methods do not.

JEPAAI summary
2026-09-30 · arXiv:2609.35545

Graph World Models for Constrained Epidemic Policy Planning

Yiqi Su, Rashed Shelim, Lingyi Wang et al.

EpiMind is a graph world-model framework for multi-region epidemic policy planning under shared-resource constraints. It uses a graph-factored recurrent state-space model (GF-RSSM) to generate joint policy-conditioned rollouts, and a graph-temporal ADMM (GT-ADMM) planner to coordinate region interventions, enforce per-period feasibility via projection, and evaluate temporal specifications on projected actions.

DrivingAI summary
2026-09-30 · arXiv:2609.35603

Control-Geometry Straightening for Sampling-Based Latent Planning

Ziang Fu, Ning Ning

The authors introduce Control-Geometry Straightening (CGS), a single auxiliary loss for joint-embedding latent world models. CGS uses only local pixel–action transitions to match pairwise cosine similarities among actions to those among corresponding latent differences, shaping “control geometry” for sampling-efficient latent planning. Under linear-dynamics, they analyze how this relates to temporal straightening, finite-budget guarantees for MPPI, and convergence results for gradient descent; experiments show large success-rate gains over LeWM and LeWM+TS.

Latent DynamicsAI summary
2026-09-30 · arXiv:2609.35811

Lookahead-R: Budget-Aware Tool Retrieval via Execution-Centric Planning

Zongze Wu, Yani Guo, Runnan Li

Lookahead-R reframes tool retrieval as a resource-constrained sequential decision problem. It trains an execution-aware surrogate world model to predict tool execution success, latency cost, and semantic utility without invoking real APIs, then uses a cost-sensitive, uncertainty-guided Monte Carlo Tree Search to select tools under latency budgets. On ToolBench I3, the authors report NDCG@5 = 91.40%, outperforming ToolGen (90.16%) by 1.24%.

DrivingAI summary
2026-09-30 · arXiv:2609.35916

VehicleArena: A Realistic Urban Environment for Multi-Agent Driving

Jie Yang, Jiajun Chen, Jiazheng Zhou et al.

VehicleArena is a high-fidelity 3D urban-driving benchmark for independently operating multi-agent driving with passenger requests and cockpit/cabin actions in a single closed loop. It provides 112 held-out evaluation tasks (80 single-agent, 32 multi-agent). The authors report that the highest arrival rates across nine evaluated models are only 65.0% (single-agent) and 65.6% (multi-agent), and driving policies can reduce surrounding vehicles’ arrival rates versus SUMO’s native traffic controller.

DrivingAI summary
2026-09-30 · arXiv:2609.35936

Embodied Semantic Communication for Collective Autonomous Agents: A Tutorial on Representation, Wireless Delivery, and Closed-Loop Coordination

Yizheng Huang, Wensheng Lin, Lixin Li et al.

This tutorial proposes embodied semantic communication (ESC) for collective autonomous agents: an explicit wireless link carries action-oriented embodied semantics (multimodal perceptual states, intrinsic kinematic/body capabilities, and collaborative intents) so heterogeneous receivers can parse, align, and ground them in local motor control within a closed-loop perception–action cycle. The paper also outlines system characteristics, environment-constrained pathways, supporting mathematical tools, and open challenges.

DrivingAI summary
2026-09-30 · arXiv:2609.36012

In-Context Learning for Robots: Methods and Applications

Haojian Huang, Zexi Li, Junhao Guo et al.

The authors review in-context learning (ICL) for robots, where demonstrations and interaction change deployed behavior without updating neural parameters during deployment. They organize prior work into four interfaces connecting context to execution: context-conditioned policies, geometric demonstration transfer, world-model-based control, and skill- and agent-based execution, analyzing how training, correspondence, and memory determine whether taught requirements transfer under changing objects, environments, and execution conditions.

Latent DynamicsAI summary
2026-09-30 · arXiv:2609.36171

SkillWeaver: Agentic Exploration over Neural Interaction Skills for Scalable Robot Data Generation

He Zhu, Lusen Zhao, Kwan Man Cheng et al.

SkillWeaver is an agentic framework for scalable robot data generation. Given a task and a simulated environment, a VLM agent reasons about what to do next, invokes and parameterizes Neural Interaction Skills (NIS)—reusable closed-loop policies for contact-rich manipulation—and uses verifier-guided tree search with reflection and experience memory to discover and verify long-horizon behaviors. It scales to 39.1K demonstrations across 14.1K scenes and distills them into visuomotor policies for generalization and sim-to-real transfer.

RoboticsAI summary
2026-09-30 · arXiv:2609.36227

One-Step Next-Latent Prediction Is Not a World Model

Shitong Wang, Zhongang Cai, Yuzhou Hong

The authors argue that next-latent (one-step) prediction is not, by itself, a world model. They show that one-step regression identifies only a conditional mean (a point), while multi-step open-loop error depends on the residual innovation covariance and grows with horizon after perfect one-step fit. They analyze cases with nonlinear means and non-injective observations, and show that an isotropy penalty depends only on the embedding marginal and has zero transition gradient.

Latent DynamicsAI summary
2026-09-30 · arXiv:2609.36305

Bilinear World Models: Learning Representations with Structured Dynamics for Efficient Control

Antonio Pariente, Ignacio Boero, Nikolai Matni et al.

The authors propose Bilinear World Models, a JEPA-style world model that restricts latent dynamics to a bilinear parameterization (with structured normalization). The structure enforces action recoverability and prevents representation collapse by construction. Using a gradient-based Gauss–Newton planner that exploits differentiable dynamics, they report substantially faster planning (two to three orders of magnitude) while matching or improving control success on 2D/3D control tasks, including longer-horizon and real-time control.

JEPAAI summary
2026-09-30 · arXiv:2609.36323

Towards an AI Software Factory for Data Systems

Anna Pavlenko, Bogdan Crivat, Brandon Haynes et al.

The authors propose an AI Software Factory for Data Systems that automates SDLC stages (Targeting, Coding, Reviewing, Ops) using a World Model and an evolutionary coding system (Darwin) wrapped in a self-improvement loop. They report scaled Microsoft deployments (tens of repositories) yielding 3× engineering efficiency vs agentic coding and up to 22× token efficiency, plus OSS and production evidence.

DrivingAI summary
2026-09-30 · arXiv:2609.36333

ATLAS: Aligned Transport of Latent Structure for Reliable World Model Planning

Ke Fang, Yupu Yao, Lu Cheng

ATLAS addresses a gap in latent world-model planning: regularizing only the latent marginal (anti-collapse) does not guarantee preservation of the relational geometry needed for goal-conditioned action selection. ATLAS transfers normalized pairwise structure from an informative encoder representation to the planning latent and calibrates the planning latent’s marginal using Wasserstein embedding matching (WEMReg). Instantiated in LeWM, it improves mean goal-reaching success, especially on higher-novelty TwoRoom episodes.

Latent DynamicsAI summary
2026-09-30 · arXiv:2609.36366

Cross-attention encoding models reveal dynamic spatiotemporal routing across human higher visual cortex

Iishaan Inabathini, Margaret M. Henderson

The authors extend cross-attention fMRI encoding models from static images to naturalistic video. They use per-parcel joint spatiotemporal cross-attention (joint over space and time) over V-JEPA-2 video tokens, fitting to BOLD Responses from short video clips. Joint routing yields higher held-out prediction accuracy across higher visual regions and produces interpretable attention maps that track moving objects and differ across category-selective networks.

Latent DynamicsAI summary
2026-09-30 · arXiv:2609.36413

One from Infinity: Actualizing Futures from Pretrained World Models into Robot Actions

Bang Du, Yichen Xie, Shuqi Zhao et al.

The authors propose RoboActualizer to turn a pretrained video world model into a robot policy by “actualizing” a single task-conditioned future. A frozen V-JEPA 2.1 encoder provides a prior over plausible future latents; a tiny flow-matching head with two DiT experts learns task-conditioned selection and action realization. The approach is trained on cached latents and deployed with low latency for real-time control.

RoboticsAI summary
2026-09-30 · arXiv:2609.36416

FineART: Fine-grained Annotated Robotic Trajectory Dataset and Vision-Language-Action Model for Bimanual Manipulation

Jade Choghari, Pepijn Kooijmans, Mansi Agarwal et al.

FineART is a densely annotated bimanual manipulation dataset (40,543 episodes, 1,718 hours, 533,913 subtasks across 151 tasks). The authors introduce FineART-VLA, a vision-language-action policy that predicts its own next subtask and executes continuous actions. Dense subtask training improves success (32.0%→100.0% on spatial disambiguation; 16.0%→76.0% on a long-horizon task). Mid-training enables data-efficient cross-robot transfer, and the authors open-source dataset, model weights, and training code.

RoboticsAI summary
2026-09-30 · arXiv:2609.36438

World4Scorer: Outcome-Grounded World Modeling for Autonomous Driving

Jieyuan Pei, Meiyi Lu, Sining Ang et al.

World4Scorer is an outcome-grounded, trajectory-conditioned JEPA-style world modeling framework for generate-and-select autonomous driving planners. A shared predictor maps each candidate plan to a latent state; simulator outcome labels supervise all candidates, while the observed future of the executed trajectory anchors shared parameters. Inertial re-ranking keeps consecutive selections consistent. The authors report state-of-the-art NAVSIM-v2 results and improved closed-loop Bench2Drive and manipulation on OGBench-Cube with a frozen LeWM world model.

DrivingAI summary
2026-09-30 · arXiv:2609.36454

DynamicHOI: Coupled Dynamics for Physics-aware HOI Reconstruction

Wenliang Guo, Zhanbo Huang, Yu Kong

DynamicHOI reconstructs hand-object interaction (HOI) trajectories from monocular RGB video using geometry-grounded diffusion refinement plus coupled hand-object physics. It derives hand generalized forces via articulated inverse dynamics and object wrenches via Newton–Euler dynamics, couples them through contact-force transfer, and recovers active hand actuation regularized by a probabilistic prior to suppress mechanically implausible motion.

RoboticsAI summary
2026-09-30 · arXiv:2609.36471

Staircase Policy: Streaming Inference for World-Action Models with Large Action Chunks

Guoheng Sun, Chen Chen, Jin Wang et al.

Staircase Policy (S-WAM) is a streaming inference/training framework for World-Action Models with large action chunks. It turns a flow-matching VLA into a JEPA-style WAM and partitions a large action chunk into sub-chunks at staggered denoising stages, executing near-term actions early and continuously refining unexecuted actions by refreshing a lightweight future-latent predictor at each sub-chunk boundary.

DrivingAI summary
2026-09-30 · arXiv:2609.36531

Foresight at the Event Boundary: Evaluating Physical Prediction in Video World Models

Estela Monserrat Arriaga Santana (National Autonomous University of Mexico), Julian Rosas Scull (National Autonomous University of Mexico), Eh\'ecatl Sacamch'en N\'u\~nez Rico (National Autonomous University of Mexico) et al.

The authors evaluate physical prediction in video world models at “event boundaries,” where release/impact is shown but its consequence is withheld. Using 62 controlled real free-fall recordings (124 event-anchored clips) and human trajectory prediction, they find Runway and Veo often produce release and impact events above 93% but with late onset, while Cosmos-Predict-2.5 and MAGI-1 frequently suppress measurable consequences. They also find that plausible timing does not guarantee physically consistent motion.