8 citations · 14 across the 24 of their papers we have counts for
24 papers
World-Value-Action Model: Implicit Planning for Vision-Language-Action Systems
Runze Li, Hongyin Zhang, Junxi Jin +5
Vision-Language-Action (VLA) models have emerged as a promising paradigm for building embodied agents that ground perception and language into action. However, most existing approa…
NFPO: Stabilized Policy Optimization of Normalizing Flow for Robotic Policy Learning
Diyuan Shi, Yiqi Tang, Zifeng Zhuang +1
Deep Reinforcement Learning (DRL) has experienced significant advancements in recent years and has been widely used in many fields. In DRL-based robotic policy learning, however, c…
MMaDA-VLA: Large Diffusion Vision-Language-Action Model with Unified Multi-Modal Instruction and Generation
Yang Liu, Pengxiang Ding, Tengyue Jiang +10
Vision-Language-Action (VLA) models map visual observations and natural-language instructions to robot actions; however, hierarchical and autoregressive paradigms often incur archi…
HiF-VLA: Hindsight, Insight and Foresight through Motion Representation for Vision-Language-Action Models
Minghui Lin, Pengxiang Ding, Shu Wang +7
Vision-Language-Action (VLA) models have recently enabled robotic manipulation by grounding visual and linguistic cues into actions. However, most VLAs assume the Markov property,…
Boundary-to-Region Supervision for Offline Safe Reinforcement Learning
Huikang Su, Dengyun Peng, Zifeng Zhuang +4
Offline safe reinforcement learning aims to learn policies that satisfy predefined safety constraints from static datasets. Existing sequence-model-based methods condition action g…
VARD: Efficient and Dense Fine-Tuning for Diffusion Models with Value-based RL
Fengyuan Dai, Zifeng Zhuang, Yufei Huang +4
Diffusion models have emerged as powerful generative tools across various domains, yet tailoring pre-trained models to exhibit specific desirable properties remains challenging. Wh…
Closing the Gap between TD Learning and Supervised Learning with -Conditioned Maximization
Xing Lei, Zifeng Zhuang, Shentao Yang +6
Recently, supervised learning (SL) methodology has emerged as an effective approach for offline reinforcement learning (RL) due to their simplicity, stability, and efficiency. Howe…
Efficient Online RL Fine Tuning with Offline Pre-trained Policy Only
Wei Xiao, Jiacheng Liu, Zifeng Zhuang +3
Improving the performance of pre-trained policies through online reinforcement learning (RL) is a critical yet challenging topic. Existing online RL fine-tuning methods require con…
ReinboT: Amplifying Robot Visual-Language Manipulation with Reinforcement Learning
Hongyin Zhang, Zifeng Zhuang, Han Zhao +3
Vision-Language-Action (VLA) models have shown great potential in general robotic decision-making tasks via imitation learning. However, the variable quality of training data often…
TDMPBC: Self-Imitative Reinforcement Learning for Humanoid Robot Control
Zifeng Zhuang, Diyuan Shi, Runze Suo +5
Complex high-dimensional spaces with high Degree-of-Freedom and complicated action spaces, such as humanoid robots equipped with dexterous hands, pose significant challenges for re…
Q-WSL: Optimizing Goal-Conditioned RL with Weighted Supervised Learning via Dynamic Programming
Xing Lei, Xuetao Zhang, Zifeng Zhuang +1
A novel class of advanced algorithms, termed Goal-Conditioned Weighted Supervised Learning (GCWSL), has recently emerged to tackle the challenges posed by sparse rewards in goal-co…
Reinformer: Max-Return Sequence Modeling for Offline RL
Zifeng Zhuang, Dengyun Peng, Jinxin Liu +2
As a data-driven paradigm, offline reinforcement learning (RL) has been formulated as sequence modeling that conditions on the hindsight information including returns, goal or futu…
DIDI: Diffusion-Guided Diversity for Offline Behavioral Generation
Jinxin Liu, Xinghong Guo, Zifeng Zhuang +1
In this paper, we propose a novel approach called DIffusion-guided DIversity (DIDI) for offline behavioral generation. The goal of DIDI is to learn a diverse set of skills from a m…
Imitating from auxiliary imperfect demonstrations via Adversarial Density Weighted Regression
Ziqi Zhang, Zifeng Zhuang, Jingzehua Xu +4
We propose a novel one-step supervised imitation learning (IL) framework called Adversarial Density Regression (ADR). This IL framework aims to correct the policy learned on unknow…
Context-Former: Stitching via Latent Conditioned Sequence Modeling
Ziqi Zhang, Jingzehua Xu, Jinxin Liu +4
Offline reinforcement learning (RL) algorithms can learn better decision-making compared to behavior policies by stitching the suboptimal trajectories to derive more optimal ones.…
A dynamical clipping approach with task feedback for Proximal Policy Optimization
Ziqi Zhang, Jingzehua Xu, Zifeng Zhuang +4
Proximal Policy Optimization (PPO) has been broadly applied to robotics learning, showcasing stable training performance. However, the fixed clipping bound setting may limit the pe…
Unlock Reliable Skill Inference for Quadruped Adaptive Behavior by Skill Graph
Hongyin Zhang, Diyuan Shi, Zifeng Zhuang +6
Developing robotic intelligent systems that can adapt quickly to unseen wild situations is one of the critical challenges in pursuing autonomous robotics. Although some impressive…
Improving Offline-to-Online Reinforcement Learning with Q Conditioned State Entropy Exploration
Ziqi Zhang, Xiao Xiong, Zifeng Zhuang +2
Studying how to fine-tune offline reinforcement learning (RL) pre-trained policy is profoundly significant for enhancing the sample efficiency of RL algorithms. However, directly f…
STRAPPER: Preference-based Reinforcement Learning via Self-training Augmentation and Peer Regularization
Yachen Kang, Li He, Jinxin Liu +2
Preference-based reinforcement learning (PbRL) promises to learn a complex reward function with binary human preference. However, such human-in-the-loop formulation requires consid…
Beyond OOD State Actions: Supported Cross-Domain Offline Reinforcement Learning
Jinxin Liu, Ziqi Zhang, Zhenyu Wei +4
Offline reinforcement learning (RL) aims to learn a policy using only pre-collected and fixed data. Although avoiding the time-consuming online interactions in RL, it poses challen…
CEIL: Generalized Contextual Imitation Learning
Jinxin Liu, Li He, Yachen Kang +3
In this paper, we present \textbf{C}ont\textbf{E}xtual \textbf{I}mitation \textbf{L}earning~(CEIL), a general and broadly applicable algorithm for imitation learning (IL). Inspired…
Design from Policies: Conservative Test-Time Adaptation for Offline Policy Optimization
Jinxin Liu, Hongyin Zhang, Zifeng Zhuang +3
In this work, we decouple the iterative bi-level offline RL (value estimation and policy extraction) from the offline training phase, forming a non-iterative bi-level paradigm and…
RotoGBML: Towards Out-of-Distribution Generalization for Gradient-Based Meta-Learning
Min Zhang, Zifeng Zhuang, Zhitao Wang +2
Gradient-based meta-learning (GBML) algorithms are able to fast adapt to new tasks by transferring the learned meta-knowledge, while assuming that all tasks come from the same dist…
Behavior Proximal Policy Optimization
Zifeng Zhuang, Kun Lei, Jinxin Liu +2
Offline reinforcement learning (RL) is a challenging setting where existing off-policy actor-critic methods perform poorly due to the overestimation of out-of-distribution state-ac…