홈 › Papers

LaMP: Learning Vision-Language-Action Policy with 3D Scene Flow as Latent Motion Prior

2026-03-26 · Xinkai Wang, Chenyi Wang, Yifu Xu, Mingzhe Ye, Fucheng Zhang, Jialin Tian, Xinyu Zhan, Lifeng Zhu, Cewu Lu, Lixin Yang arxiv

We introduce \textbf{LaMP}, a dual-expert Vision-Language-Action framework that embeds dense 3D scene flow as a latent motion prior for robotic manipulation.Existing VLA models regress actions directly from 2D semantic visual features, forcing them to learn complex 3D physical interactions implicitly.This implicit learning strategy degrades under unfamiliar spatial dynamics.LaMP addresses this limitation by aligning a flow-matching \emph{Motion Expert} with a policy-predicting \emph{Action Expert} through gated cross-attention.Specifically, the Motion Expert generates a one-step partially denoised 3D scene flow, and its hidden states condition the Action Expert without full multi-step reconstruction.We evaluate LaMP on the LIBERO, LIBERO-Plus, and SimplerEnv-WidowX simulation benchmarks as well as real-world experiments.LaMP consistently outperforms evaluated VLA baselines across LIBERO, LIBERO-Plus, and SimplerEnv-WidowX benchmarks, achieving the highest reported average success rates under the same training budgets. On LIBERO-Plus OOD perturbations, LaMP shows improved robustness with an average 9.7\% gain over the strongest prior baseline.Our project page is available at https://summerwxk.github.io/lamp-project-page/.

📄 PDF Abstract BibTeX arXiv:2603.25399

Code (0)

등록된 구현이 없습니다.

CLAMP: Contrastive Learning for 3D Multi-View Action-Conditioned Robotic Manipulation Pretraining

2026-01-31 · I-Chun Arthur Liu, Krzysztof Choromanski, Sandy Huang, Connor Schenck arxiv

Leveraging pre-trained 2D image representations in behavior cloning policies has achieved great success and has become a standard approach for robotic manipulation. However, such representations fail to capture the 3D sp…

Contrastive Learning Point Clouds

A-LAMP: Agentic LLM-Based Framework for Automated MDP Modeling and Policy Generation

2025-12-12 · Hong Je-Gal, Chan-Bin Yi, Hyun-Suk Lee arxiv

Applying reinforcement learning (RL) to real-world tasks requires converting informal descriptions into a formal Markov decision process (MDP), implementing an executable environment, and training a policy agent. Automat…

Reinforcement Learning

LAMP: Leveraging Language Prompts for Multi-person Pose Estimation

2023-07-21 · Shengnan Hu, Ce Zheng, Zixiang Zhou, Chen Chen 외

Human-centric visual understanding is an important desideratum for effective human-robot interaction. In order to navigate crowded public places, social robots must be able to interpret the activity of the surrounding hu…

Language Modeling Language Modelling Multi-Person Pose Estimation Navigate+1

Track4Action: Distilling World-Centric 3D Tracker into Vision-Language-Action Policies

2026-08-04 · Chenyi Wang, Xinkai Wang, Bokai Lin, Jialin Tian 외 arxiv

Action labels tell a vision-language-action (VLA) policy which robot commands to imitate, but not how those commands change the 3D world. The aligned demonstration clip contains this missing supervision because its $K$ f…

Language Reward Modulation for Pretraining Reinforcement Learning

2023-08-23 · Ademi Adeniji, Amber Xie, Carmelo Sferrazza, Younggyo Seo 외

Using learned reward functions (LRFs) as a means to solve sparse-reward reinforcement learning (RL) tasks has yielded some steady progress in task-complexity through the years. In this work, we question whether today's L…

reinforcement-learning Reinforcement Learning Reinforcement Learning (RL)Robot Manipulation