Seeing Beyond the Next Step: World-Model-Guided Human-Like Navigation in Multi-Agent Scenes

原文：Seeing Beyond the Next Step: World-Model-Guided Human-Like Navigation in Multi-Agent Scenes

作者：Seong-Eun Hong, JuYeong Hwang, Jinhyun Kim, Hanyoung Jang, HyeongYeop Kang

来源：arXiv cs.RO（机器人）

正文

Computer Science > Graphics

arXiv:2611.07730v1 (cs)

[Submitted on 5 Nov 2026]

Title:Seeing Beyond the Next Step: World-Model-Guided Human-Like Navigation in Multi-Agent Scenes

Authors:Seong-Eun Hong, JuYeong Hwang, Jinhyun Kim, Hanyoung Jang, HyeongYeop Kang

View a PDF of the paper titled Seeing Beyond the Next Step: World-Model-Guided Human-Like Navigation in Multi-Agent Scenes, by Seong-Eun Hong and 4 other authors

View PDF

HTML (experimental)

Abstract:Selecting a next motion in a populated scene is not just a matter of avoiding what is currently in the way: an action can be locally feasible yet drift onto the road instead of the crosswalk, cut into a stream of crossing pedestrians, or bottleneck a narrow passage. These failure modes are invisible from the current state alone, and acting on them requires reasoning about what each candidate action will cause. World models offer exactly this, namely action-conditioned predictions of the future, but rolling them out at every decision step is a cost deployable navigation systems cannot pay, and the field has largely traded foresight for speed. We argue this is a false dichotomy. The natural place for a world model in human-aware navigation is at training time, as a preference oracle that distills foresight into a fast policy, rather than at inference time, as an online planner. We pair a discrete visual action prior with two task-aligned world models, one for nearby social dynamics and one for future scene semantics, that score candidate actions by their predicted consequences and refine the policy through group-relative preference distillation. At deployment, both world models are removed and the policy acts in a single forward pass. Across five held-out evaluation scenarios, this yields navigation that is simultaneously safer, more socially and group-coherent, and more goal-directed than strong reactive and learned baselines, with naive observers rating the resulting motion as more human-like. Taken together, our findings point to a different approach for bringing world-model foresight into deployable agents: not accelerating rollouts, but amortizing them into the policy so that the cost of looking ahead is confined to training. Project page: this https URL

Comments:

Accepted to NeurIPS 2026. 37 pages, 13 figures, 12 tables

Subjects:

Graphics (cs.GR); Artificial Intelligence (cs.AI); Computer Vision and Pattern Recognition (cs.CV); Robotics (cs.RO)

Cite as:

arXiv:2611.07730 [cs.GR]

(or

arXiv:2611.07730v1 [cs.GR] for this version)

https://doi.org/10.48550/arXiv.2611.07730

Focus to learn more

arXiv-issued DOI via DataCite (pending registration)

主题

cs.GR · 人工智能 · 计算机视觉 · 机器人

由「前沿雷达」于 2026-11-05 采集。正文取自原文页面，已保留出处链接。标题与正文版权归原作者所有。