paper-with-me

홈 › Papers

Act2Goal: From World Model To General Goal-conditioned Policy

2025-12-29 · Pengfei Zhou, Liliang Chen, Shengcong Chen, Di Chen, Wenzhi Zhao, Rongjun Jin, Guanghui Ren, Jianlan Luo arxiv

Specifying robotic manipulation tasks in a manner that is both expressive and precise remains a central challenge. While visual goals provide a compact and unambiguous task specification, existing goal-conditioned policies often struggle with long-horizon manipulation due to their reliance on single-step action prediction without explicit modeling of task progress. We propose Act2Goal, a general goal-conditioned manipulation policy that integrates a goal-conditioned visual world model with multi-scale temporal control. Given a current observation and a target visual goal, the world model generates a plausible sequence of intermediate visual states that captures long-horizon structure. To translate this visual plan into robust execution, we introduce Multi-Scale Temporal Hashing (MSTH), which decomposes the imagined trajectory into dense proximal frames for fine-grained closed-loop control and sparse distal frames that anchor global task consistency. The policy couples these representations with motor control through end-to-end cross-attention, enabling coherent long-horizon behavior while remaining reactive to local disturbances. Act2Goal achieves strong zero-shot generalization to novel objects, spatial layouts, and environments. We further enable reward-free online adaptation through hindsight goal relabeling with LoRA-based finetuning, allowing rapid autonomous improvement without external supervision. Real-robot experiments demonstrate that Act2Goal improves success rates from 30% to 90% on challenging out-of-distribution tasks within minutes of autonomous interaction, validating that goal-conditioned world models with multi-scale temporal control provide structured guidance necessary for robust long-horizon manipulation. Project page: https://act2goal.github.io/

📄 PDF Abstract BibTeX arXiv:2512.23541

Code (0)

등록된 구현이 없습니다.

Tasks

Zero-shot Generalization

Similar Papers 제목 키워드 기반

Learn Goal-Conditioned Policy with Intrinsic Motivation for Deep Reinforcement Learning

2021-04-11 · Jinxin Liu, Donglin Wang, Qiangxing Tian, Zhengyu Chen

It is of significance for an agent to learn a widely applicable and general-purpose policy that can achieve diverse goals including images and text descriptions. Considering such perceptually-specific goals, the frontier…

Deep Reinforcement Learningreinforcement-learningReinforcement Learning (RL)

Planning to Practice: Efficient Online Fine-Tuning by Composing Goals in Latent Space

2022-05-17 · Kuan Fang, Patrick Yin, Ashvin Nair, Sergey Levine

General-purpose robots require diverse repertoires of behaviors to complete challenging tasks in real-world unstructured environments. To address this issue, goal-conditioned reinforcement learning aims to acquire polici…

reinforcement-learningReinforcement Learning (RL)

Goal-Conditioned Imitation Learning using Score-based Diffusion Policies

2023-04-05 · Moritz Reuss, Maximilian Li, Xiaogang Jia, Rudolf Lioutikov

We propose a new policy representation based on score-based diffusion models (SDMs). We apply our new policy representation in the domain of Goal-Conditioned Imitation Learning (GCIL) to learn general-purpose goal-specif…

DenoisingImitation LearningRobot Manipulation

GOPlan: Goal-conditioned Offline Reinforcement Learning by Planning with Learned Models

2023-10-30 · Mianchu Wang, Rui Yang, Xi Chen, Hao Sun 외

Offline Goal-Conditioned RL (GCRL) offers a feasible paradigm for learning general-purpose policies from diverse and multi-task offline datasets. Despite notable recent progress, the predominant offline GCRL methods, mai…

Generative Adversarial Networkreinforcement-learning

Policy-Guided World Model Planning for Language-Conditioned Visual Navigation

2026-03-26 · Amirhosein Chahe, Lifeng Zhou arxiv

Navigating to a visually specified goal given natural language instructions remains a fundamental challenge in embodied AI. Existing approaches either rely on reactive policies that struggle with long-horizon planning, o…

Visual Navigation