Recent advances in learning decision-making policies can largely be attributed to training expressive policy models, largely via imitation learning. While imitation learning discards non-expert data, reinforcement learning (RL) can still learn from suboptimal data. However, instantiating RL training of a new policy class often presents a different challenge: most deep RL machinery is co-developed with assumptions on the policy class and backbone, resulting in poor performance when the policy class changes. For instance, SAC utilizes a low-variance reparameterization policy gradient for Gaussian policies, but this is unstable for diffusion policies and intractable for autoregressive categorical policies. To address this issue, we develop an offline RL and online fine-tuning approach called policy-agnostic RL (PA-RL) that can effectively train multiple policy classes, with varying architectures and sizes. We build off the basic idea that a universal supervised learning loss can replace the policy improvement step in RL, as long as it is applied on "optimized" actions. To obtain these optimized actions, we first sample multiple actions from a base policy, and run global optimization (i.e., re-ranking multiple action samples using the Q-function) and local optimization (i.e., running gradient steps on an action sample) to maximize the critic on these candidates. PA-RL enables fine-tuning diffusion and transformer policies with either autoregressive tokens or continuous action outputs, at different sizes, entirely via actor-critic RL. Moreover, PA-RL improves the performance and sample-efficiency by up to 2 times compared to existing offline RL and online fine-tuning methods. We show the first result that successfully fine-tunes OpenVLA, a 7B generalist robot policy, autonomously with Cal-QL, an online RL fine-tuning algorithm, improving from 40% to 70% in the real world in 40 minutes.
Large language models (LLMs) have made dialogue one of the central modes in human-machine interaction, leading to the vast amounts of conversation logs and increasing demand for dialogue generation. The dialogue's life-cycle spans from the $\textit{Prelude}$ through the $\textit{Interlocution}$ to the $\textit{Epilogue}$, encompassing rich dialogue elements. Despite the large volumes of dialogue-related studies, there is a lack of benchmark that encompasses comprehensive dialogue elements, which hinders precise modeling, generation and systematic evaluation. To bridge this gap, in this paper, we introduce a new research task $\textbf{D}$ialogue $\textbf{E}$lement $\textbf{MO}$deling, including $\textit{Element Awareness}$ and $\textit{Dialogue Agent Interaction}$, and propose a novel benchmark, $\textbf{DEMO}$, designed for a comprehensive dialogue modeling and assessment. On this basis, we further build the DEMO agent with the adept ability to model dialogue elements via imitation learning. Extensive experiments on DEMO indicate that current representative LLMs still have considerable potential for enhancement, and our DEMO agent performs well in both dialogue element modeling and out-of-domain tasks.
Collaboration is a cornerstone of society. In the real world, human teammates make use of multi-sensory data to tackle challenging tasks in ever-changing environments. It is essential for embodied agents collaborating in visually-rich environments replete with dynamic interactions to understand multi-modal observations and task specifications. To evaluate the performance of generalizable multi-modal collaborative agents, we present TeamCraft, a multi-modal multi-agent benchmark built on top of the open-world video game Minecraft. The benchmark features 55,000 task variants specified by multi-modal prompts, procedurally-generated expert demonstrations for imitation learning, and carefully designed protocols to evaluate model generalization capabilities. We also perform extensive analyses to better understand the limitations and strengths of existing approaches. Our results indicate that existing models continue to face significant challenges in generalizing to novel goals, scenes, and unseen numbers of agents. These findings underscore the need for further research in this area. The TeamCraft platform and dataset are publicly available at https://github.com/teamcraft-bench/teamcraft.
Federated Learning (FL) is a privacy-preserving distributed learning paradigm designed to build a highly accurate global model. In Mobile Edge IoT (MEIoT), the training and communication processes can significantly deplete the limited battery resources of devices. Existing research primarily focuses on reducing overall energy consumption, but this may inadvertently create energy consumption imbalances, leading to the premature dropout of energy-sensitive devices.To address these challenges, we propose BEFL, a joint optimization framework aimed at balancing three objectives: enhancing global model accuracy, minimizing total energy consumption, and reducing energy usage disparities among devices. First, taking into account the communication constraints of MEIoT and the heterogeneity of devices, we employed the Sequential Least Squares Programming (SLSQP) algorithm for the rational allocation of communication resources. Based on this, we introduce a heuristic client selection algorithm that combines cluster partitioning with utility-driven approaches to alleviate both the total energy consumption of all devices and the discrepancies in energy usage.Furthermore, we utilize the proposed heuristic client selection algorithm as a template for offline imitation learning during pre-training, while adopting a ranking-based reinforcement learning approach online to further boost training efficiency. Our experiments reveal that BEFL improves global model accuracy by 1.6\%, reduces energy consumption variance by 72.7\%, and lowers total energy consumption by 28.2\% compared to existing methods. The relevant code can be found at \href{URL}{https://github.com/juzehao/BEFL}.
Virtual models of human gait, or digital twins, offer a promising solution for studying mobility without the need for labor-intensive data collection. However, challenges such as the sim-to-real gap and limited adaptability to diverse walking conditions persist. To address these, we developed and validated a framework to create a skeletal humanoid agent capable of adapting to varying walking speeds while maintaining biomechanically realistic motions. The framework combines a synthetic data generator, which produces biomechanically plausible gait kinematics from open-source biomechanics data, and a training system that uses adversarial imitation learning to train the agent's walking policy. We conducted comprehensive analyses comparing the agent's kinematics, synthetic data, and the original biomechanics dataset. The agent achieved a root mean square error of 5.24 +- 0.09 degrees at varying speeds compared to ground-truth kinematics data, demonstrating its adaptability. This work represents a significant step toward developing a digital twin of human locomotion, with potential applications in biomechanics research, exoskeleton design, and rehabilitation.
Offline reinforcement learning (RL) seeks to learn optimal policies from static datasets without interacting with the environment. A common challenge is handling multi-modal action distributions, where multiple behaviours are represented in the data. Existing methods often assume unimodal behaviour policies, leading to suboptimal performance when this assumption is violated. We propose Weighted Imitation Learning on One Mode (LOM), a novel approach that focuses on learning from a single, promising mode of the behaviour policy. By using a Gaussian mixture model to identify modes and selecting the best mode based on expected returns, LOM avoids the pitfalls of averaging over conflicting actions. Theoretically, we show that LOM improves performance while maintaining simplicity in policy learning. Empirically, LOM outperforms existing methods on standard D4RL benchmarks and demonstrates its effectiveness in complex, multi-modal scenarios.
Because imitation learning relies on human demonstrations in hard-to-simulate settings, the inclusion of force control in this method has resulted in a shortage of training data, even with a simple change in speed. Although the field of data augmentation has addressed the lack of data, conventional methods of data augmentation for robot manipulation are limited to simulation-based methods or downsampling for position control. This paper proposes a novel method of data augmentation that is applicable to force control and preserves the advantages of real-world datasets. We applied teaching-playback at variable speeds as real-world data augmentation to increase both the quantity and quality of environmental reactions at variable speeds. An experiment was conducted on bilateral control-based imitation learning using a method of imitation learning equipped with position-force control. We evaluated the effect of real-world data augmentation on two tasks, pick-and-place and wiping, at variable speeds, each from two human demonstrations at fixed speed. The results showed a maximum 55% increase in success rate from a simple change in speed of real-world reactions and improved accuracy along the duration/frequency command by gathering environmental reactions at variable speeds.
Imitation Learning presents a promising approach for learning generalizable and complex robotic skills. The recently proposed Diffusion Policy generates robot action sequences through a conditional denoising diffusion process, achieving state-of-the-art performance compared to other imitation learning methods. This paper summarizes five key components of Diffusion Policy: 1) observation sequence input; 2) action sequence execution; 3) receding horizon; 4) U-Net or Transformer network architecture; and 5) FiLM conditioning. By conducting experiments across ManiSkill and Adroit benchmarks, this study aims to elucidate the contribution of each component to the success of Diffusion Policy in various scenarios. We hope our findings will provide valuable insights for the application of Diffusion Policy in future research and industry.
Deep neural network (DNN)-based policy models like vision-language-action (VLA) models are transformative in automating complex decision-making across applications by interpreting multi-modal data. However, scaling these models greatly increases computational costs, which presents challenges in fields like robot manipulation and autonomous driving that require quick, accurate responses. To address the need for deployment on resource-limited hardware, we propose a new quantization framework for IL-based policy models that fine-tunes parameters to enhance robustness against low-bit precision errors during training, thereby maintaining efficiency and reliability under constrained conditions. Our evaluations with representative robot manipulation for 4-bit weight-quantization on a real edge GPU demonstrate that our framework achieves up to 2.5x speedup and 2.5x energy savings while preserving accuracy. For 4-bit weight and activation quantized self-driving models, the framework achieves up to 3.7x speedup and 3.1x energy saving on a low-end GPU. These results highlight the practical potential of deploying IL-based policy models on resource-constrained devices.
Today's largest foundation models have increasingly general capabilities, yet when used as agents, they often struggle with simple reasoning and decision-making tasks, even though they possess good factual knowledge of the task and how to solve it. In this paper, we present a benchmark to pressure-test these models' multimodal decision-making capabilities in the very long-context regime (up to one million tokens) and investigate whether they can learn from a large number of expert demonstrations in their context. We evaluate a wide range of state-of-the-art frontier models as policies across a battery of simple interactive decision-making tasks: playing tic-tac-toe, chess, and Atari, navigating grid worlds, solving crosswords, and controlling a simulated cheetah. We measure the performance of Claude 3.5 Sonnet, Gemini 1.5 Flash, Gemini 1.5 Pro, GPT-4o, o1-mini, and o1-preview under increasing amounts of expert demonstrations in the context $\unicode{x2013}$ from no demonstrations up to 512 full episodes, pushing these models' multimodal long-context reasoning capabilities to their limits. Across our tasks, today's frontier models rarely manage to fully reach expert performance, showcasing the difficulty of our benchmark. Presenting more demonstrations often has little effect, but some models steadily improve with more demonstrations on a few tasks. We investigate the effect of encoding observations as text or images and the impact of chain-of-thought prompting. Overall, our results suggest that even today's most capable models often struggle to imitate desired behavior by generalizing purely from in-context demonstrations. To help quantify the impact of other approaches and future innovations aiming to tackle this problem, we open source our benchmark that covers the zero-, few-, and many-shot regimes in a unified evaluation.
Humanoid robots have significant gaps in their sensing and perception, making it hard to perform motion planning in dense environments. To address this, we introduce ARMOR, a novel egocentric perception system that integrates both hardware and software, specifically incorporating wearable-like depth sensors for humanoid robots. Our distributed perception approach enhances the robot's spatial awareness, and facilitates more agile motion planning. We also train a transformer-based imitation learning (IL) policy in simulation to perform dynamic collision avoidance, by leveraging around 86 hours worth of human realistic motions from the AMASS dataset. We show that our ARMOR perception is superior against a setup with multiple dense head-mounted, and externally mounted depth cameras, with a 63.7% reduction in collisions, and 78.7% improvement on success rate. We also compare our IL policy against a sampling-based motion planning expert cuRobo, showing 31.6% less collisions, 16.9% higher success rate, and 26x reduction in computational latency. Lastly, we deploy our ARMOR perception on our real-world GR1 humanoid from Fourier Intelligence. We are going to update the link to the source code, HW description, and 3D CAD files in the arXiv version of this text.
Batch planning is increasingly necessary to quickly produce diverse and quality motion plans for downstream learning applications, such as distillation and imitation learning. This paper presents Global Tensor Motion Planning (GTMP) -- a sampling-based motion planning algorithm comprising only tensor operations. We introduce a novel discretization structure represented as a random multipartite graph, enabling efficient vectorized sampling, collision checking, and search. We provide a theoretical investigation showing that GTMP exhibits probabilistic completeness while supporting modern GPU/TPU. Additionally, by incorporating smooth structures into the multipartite graph, GTMP directly plans smooth splines without requiring gradient-based optimization. Experiments on lidar-scanned occupancy maps and the MotionBenchMarker dataset demonstrate GTMP's computation efficiency in batch planning compared to baselines, underscoring GTMP's potential as a robust, scalable planner for diverse applications and large-scale robot learning tasks.
Navigating dynamic urban environments presents significant challenges for embodied agents, requiring advanced spatial reasoning and adherence to common-sense norms. Despite progress, existing visual navigation methods struggle in map-free or off-street settings, limiting the deployment of autonomous agents like last-mile delivery robots. To overcome these obstacles, we propose a scalable, data-driven approach for human-like urban navigation by training agents on thousands of hours of in-the-wild city walking and driving videos sourced from the web. We introduce a simple and scalable data processing pipeline that extracts action supervision from these videos, enabling large-scale imitation learning without costly annotations. Our model learns sophisticated navigation policies to handle diverse challenges and critical scenarios. Experimental results show that training on large-scale, diverse datasets significantly enhances navigation performance, surpassing current methods. This work shows the potential of using abundant online video data to develop robust navigation policies for embodied agents in dynamic urban settings. Project homepage is at https://ai4ce.github.io/CityWalker/.
Recent advances in imitation learning for 3D robotic manipulation have shown promising results with diffusion-based policies. However, achieving human-level dexterity requires seamless integration of geometric precision and semantic understanding. We present G3Flow, a novel framework that constructs real-time semantic flow, a dynamic, object-centric 3D semantic representation by leveraging foundation models. Our approach uniquely combines 3D generative models for digital twin creation, vision foundation models for semantic feature extraction, and robust pose tracking for continuous semantic flow updates. This integration enables complete semantic understanding even under occlusions while eliminating manual annotation requirements. By incorporating semantic flow into diffusion policies, we demonstrate significant improvements in both terminal-constrained manipulation and cross-object generalization. Extensive experiments across five simulation tasks show that G3Flow consistently outperforms existing approaches, achieving up to 68.3% and 50.1% average success rates on terminal-constrained manipulation and cross-object generalization tasks respectively. Our results demonstrate the effectiveness of G3Flow in enhancing real-time dynamic semantic feature understanding for robotic manipulation policies.
Diffusion models have demonstrated remarkable capabilities in image generation tasks, including image editing and video creation, representing a good understanding of the physical world. On the other line, diffusion models have also shown promise in robotic control tasks by denoising actions, known as diffusion policy. Although the diffusion generative model and diffusion policy exhibit distinct capabilities--image prediction and robotic action, respectively--they technically follow a similar denoising process. In robotic tasks, the ability to predict future images and generate actions is highly correlated since they share the same underlying dynamics of the physical world. Building on this insight, we introduce PAD, a novel visual policy learning framework that unifies image Prediction and robot Action within a joint Denoising process. Specifically, PAD utilizes Diffusion Transformers (DiT) to seamlessly integrate images and robot states, enabling the simultaneous prediction of future images and robot actions. Additionally, PAD supports co-training on both robotic demonstrations and large-scale video datasets and can be easily extended to other robotic modalities, such as depth images. PAD outperforms previous methods, achieving a significant 26.3% relative improvement on the full Metaworld benchmark, by utilizing a single text-conditioned visual policy within a data-efficient imitation learning setting. Furthermore, PAD demonstrates superior generalization to unseen tasks in real-world robot manipulation settings with 28.0% success rate increase compared to the strongest baseline. Project page at https://sites.google.com/view/pad-paper
Recent learning-to-imitation methods have shown promising results in planning via imitating within the observation-action space. However, their ability in open environments remains constrained, particularly in long-horizon tasks. In contrast, traditional symbolic planning excels in long-horizon tasks through logical reasoning over human-defined symbolic spaces but struggles to handle observations beyond symbolic states, such as high-dimensional visual inputs encountered in real-world scenarios. In this work, we draw inspiration from abductive learning and introduce a novel framework \textbf{AB}ductive \textbf{I}mitation \textbf{L}earning (ABIL) that integrates the benefits of data-driven learning and symbolic-based reasoning, enabling long-horizon planning. Specifically, we employ abductive reasoning to understand the demonstrations in symbolic space and design the principles of sequential consistency to resolve the conflicts between perception and reasoning. ABIL generates predicate candidates to facilitate the perception from raw observations to symbolic space without laborious predicate annotations, providing a groundwork for symbolic planning. With the symbolic understanding, we further develop a policy ensemble whose base policies are built with different logical objectives and managed through symbolic reasoning. Experiments show that our proposal successfully understands the observations with the task-relevant symbolics to assist the imitation learning. Importantly, ABIL demonstrates significantly improved data efficiency and generalization across various long-horizon tasks, highlighting it as a promising solution for long-horizon planning. Project website: \url{https://www.lamda.nju.edu.cn/shaojj/KDD25_ABIL/}.
Decision-making and planning in autonomous driving critically reflect the safety of the system, making effective planning imperative. Current imitation learning-based planning algorithms often merge historical trajectories with present observations to predict future candidate paths. However, these algorithms typically assess the current and historical plans independently, leading to discontinuities in driving intentions and an accumulation of errors with each step in a discontinuous plan. To tackle this challenge, this paper introduces LHPF, an imitation learning planner that integrates historical planning information. Our approach employs a historical intention aggregation module that pools historical planning intentions, which are then combined with a spatial query vector to decode the final planning trajectory. Furthermore, we incorporate a comfort auxiliary task to enhance the human-like quality of the driving behavior. Extensive experiments using both real-world and synthetic data demonstrate that LHPF not only surpasses existing advanced learning-based planners in planning performance but also marks the first instance of a purely learning-based planner outperforming the expert. Additionally, the application of the historical intention aggregation module across various backbones highlights the considerable potential of the proposed method. The code will be made publicly available.
Recent advances in imitation learning have shown significant promise for robotic control and embodied intelligence. However, achieving robust generalization across diverse mounted camera observations remains a critical challenge. In this paper, we introduce a video-based spatial perception framework that leverages 3D spatial representations to address environmental variability, with a focus on handling lighting changes. Our approach integrates a novel image augmentation technique, AugBlender, with a state-of-the-art monocular depth estimation model trained on internet-scale data. Together, these components form a cohesive system designed to enhance robustness and adaptability in dynamic scenarios. Our results demonstrate that our approach significantly boosts the success rate across diverse camera exposures, where previous models experience performance collapse. Our findings highlight the potential of video-based spatial perception models in advancing robustness for end-to-end robotic learning, paving the way for scalable, low-cost solutions in embodied intelligence.
This paper proposes a distributed on-orbit spacecraft assembly algorithm, where future spacecraft can assemble modules with different functions on orbit to form a spacecraft structure with specific functions. This form of spacecraft organization has the advantages of reconfigurability, fast mission response and easy maintenance. Reasonable and efficient on-orbit self-reconfiguration algorithms play a crucial role in realizing the benefits of distributed spacecraft. This paper adopts the framework of imitation learning combined with reinforcement learning for strategy learning of module handling order. A robot arm motion algorithm is then designed to execute the handling sequence. We achieve the self-reconfiguration handling task by creating a map on the surface of the module, completing the path point planning of the robotic arm using A*. The joint planning of the robotic arm is then accomplished through forward and reverse kinematics. Finally, the results are presented in Unity3D.
Imitation learning in robotics faces significant challenges in generalization due to the complexity of robotic environments and the high cost of data collection. We introduce RoCoDA, a novel method that unifies the concepts of invariance, equivariance, and causality within a single framework to enhance data augmentation for imitation learning. RoCoDA leverages causal invariance by modifying task-irrelevant subsets of the environment state without affecting the policy's output. Simultaneously, we exploit SE(3) equivariance by applying rigid body transformations to object poses and adjusting corresponding actions to generate synthetic demonstrations. We validate RoCoDA through extensive experiments on five robotic manipulation tasks, demonstrating improvements in policy performance, generalization, and sample efficiency compared to state-of-the-art data augmentation methods. Our policies exhibit robust generalization to unseen object poses, textures, and the presence of distractors. Furthermore, we observe emergent behavior such as re-grasping, indicating policies trained with RoCoDA possess a deeper understanding of task dynamics. By leveraging invariance, equivariance, and causality, RoCoDA provides a principled approach to data augmentation in imitation learning, bridging the gap between geometric symmetries and causal reasoning. Project Page: https://rocoda.github.io