Copertina

WorldCompass: Reinforcement Learning for Long-Horizon World Models

Fuente: arXiv

Salvato in:

Dettagli Bibliografici
Autori principali:	Wang, Zehan, Wang, Tengfei, Zhang, Haiyu, Zuo, Xuhui, Wu, Junta, Wang, Haoyuan, Sun, Wenqiang, Wang, Zhenwei, Cao, Chenjie, Zhao, Hengshuang, Guo, Chunchao, Zhao, Zhou
Natura:	Preprint
Pubblicazione:	2026
Soggetti:	Computer Vision and Pattern Recognition
Accesso online:	Acceder al recurso
Tags:	Aggiungi Tag Nessun Tag, puoi essere il primo ad aggiungerne!!

Documenti analoghi

WorldPlay: Towards Long-Term Geometric Consistency for Real-Time Interactive World Modeling
di: Sun, Wenqiang, et al.
Pubblicazione: (2025)

WorldStereo: Bridging Camera-Guided Video Generation and Scene Reconstruction via 3D Geometric Memories
di: Zhang, Yisu, et al.
Pubblicazione: (2026)

WorldMirror: Universal 3D World Reconstruction with Any-Prior Prompting
di: Liu, Yifan, et al.
Pubblicazione: (2025)

Voyager: Long-Range and World-Consistent Video Diffusion for Explorable 3D Scene Generation
di: Huang, Tianyu, et al.
Pubblicazione: (2025)

MoCA: Mixture-of-Components Attention for Scalable Compositional 3D Generation
di: Li, Zhiqi, et al.
Pubblicazione: (2025)

Pathwise Test-Time Correction for Autoregressive Long Video Generation
di: Xiang, Xunzhi, et al.
Pubblicazione: (2026)

FlashWorld: High-quality 3D Scene Generation within Seconds
di: Li, Xinyang, et al.
Pubblicazione: (2025)

HY-World 2.0: A Multi-Modal World Model for Reconstructing, Generating, and Simulating 3D Worlds
di: HY-World, Team, et al.
Pubblicazione: (2026)

Orient Anything: Learning Robust Object Orientation Estimation from Rendering 3D Models
di: Wang, Zehan, et al.
Pubblicazione: (2024)

Depth Anything with Any Prior
di: Wang, Zehan, et al.
Pubblicazione: (2025)

PlayerOne: Egocentric World Simulator
di: Tu, Yuanpeng, et al.
Pubblicazione: (2025)

WorldWeaver: Generating Long-Horizon Video Worlds via Rich Perception
di: Liu, Zhiheng, et al.
Pubblicazione: (2025)

HunyuanWorld 1.0: Generating Immersive, Explorable, and Interactive 3D Worlds from Words or Pixels
di: HunyuanWorld Team, et al.
Pubblicazione: (2025)

LongBench: Evaluating Robotic Manipulation Policies on Real-World Long-Horizon Tasks
di: Chen, Xueyao, et al.
Pubblicazione: (2026)

GenSpace: Benchmarking Spatially-Aware Image Generation
di: Wang, Zehan, et al.
Pubblicazione: (2025)

MagicWorld: Towards Long-Horizon Stability for Interactive Video World Exploration
di: Li, Guangyuan, et al.
Pubblicazione: (2025)

AR Forcing: Towards Long-Horizon Robot Navigation World Model
di: Yang, Yifei, et al.
Pubblicazione: (2026)

TRIP-Bench: A Benchmark for Long-Horizon Interactive Agents in Real-World Scenarios
di: Shen, Yuanzhe, et al.
Pubblicazione: (2026)

E2HiL: Entropy-Guided Sample Selection for Efficient Real-World Human-in-the-Loop Reinforcement Learning
di: Deng, Haoyuan, et al.
Pubblicazione: (2026)

FieldSeer I: Physics-Guided World Models for Long-Horizon Electromagnetic Dynamics under Partial Observability
di: Guo, Ziheng, et al.
Pubblicazione: (2025)

OmniBind: Large-scale Omni Multimodal Representation via Binding Spaces
di: Wang, Zehan, et al.
Pubblicazione: (2024)

EvolvingAgent: Curriculum Self-evolving Agent with Continual World Model for Long-Horizon Tasks
di: Feng, Tongtong, et al.
Pubblicazione: (2025)

LARM: Large Auto-Regressive Model for Long-Horizon Embodied Intelligence
di: Li, Zhuoling, et al.
Pubblicazione: (2024)

STAGE: A Stream-Centric Generative World Model for Long-Horizon Driving-Scene Simulation
di: Wang, Jiamin, et al.
Pubblicazione: (2025)

RealisMotion: Decomposed Human Motion Control and Video Generation in the World Space
di: Liang, Jingyun, et al.
Pubblicazione: (2025)

World-Ego Modeling for Long-Horizon Evolution in Hybrid Embodied Tasks
di: Lin, Zuyao, et al.
Pubblicazione: (2026)

Neural Motion Simulator: Pushing the Limit of World Models in Reinforcement Learning
di: Hao, Chenjie, et al.
Pubblicazione: (2025)

PAN: A World Model for General, Interactable, and Long-Horizon World Simulation
di: PAN Team, et al.
Pubblicazione: (2025)

World to Code: Multi-modal Data Generation via Self-Instructed Compositional Captioning and Filtering
di: Wang, Jiacong, et al.
Pubblicazione: (2024)

BranPO: Scalable Contrastive Branch Sampling for Long-Horizon Agentic Reinforcement Learning
di: Zhao, Yubao, et al.
Pubblicazione: (2026)

ODYSSEY: Open-World Quadrupeds Exploration and Manipulation for Long-Horizon Tasks
di: Wang, Kaijun, et al.
Pubblicazione: (2025)

Reinforcing VLAs in Task-Agnostic World Models
di: Wang, Yucen, et al.
Pubblicazione: (2026)

Matrix-Game 3.0: Real-Time and Streaming Interactive World Model with Long-Horizon Memory
di: Wang, Zile, et al.
Pubblicazione: (2026)

Open-World Reinforcement Learning over Long Short-Term Imagination
di: Li, Jiajian, et al.
Pubblicazione: (2024)

VLN-R1: Vision-Language Navigation via Reinforcement Fine-Tuning
di: Qi, Zhangyang, et al.
Pubblicazione: (2025)

One for All: Multi-Domain Joint Training for Point Cloud Based 3D Object Detection
di: Wang, Zhenyu, et al.
Pubblicazione: (2024)

HorizonDrive: Self-Corrective Autoregressive World Model for Long-horizon Driving Simulation
di: Zhang, Conglang, et al.
Pubblicazione: (2026)

From Visual Synthesis to Interactive Worlds: Toward Production-Ready 3D Asset Generation
di: Wu, Jiafeng, et al.
Pubblicazione: (2026)

DisCo: Towards Distinct and Coherent Visual Encapsulation in Video MLLMs
di: Zhao, Jiahe, et al.
Pubblicazione: (2025)

ClinicalLab: Aligning Agents for Multi-Departmental Clinical Diagnostics in the Real World
di: Yan, Weixiang, et al.
Pubblicazione: (2024)