arXiv subjects

Kexin Zheng

Publications and source records attributed to Kexin Zheng.

11 recordsLinked to original sources

arXiv2026

Dichotomous Diffusion Policy Optimization

Diffusion-based policies have gained growing popularity in solving a wide range of decision-making tasks due to their superior expressiveness and controllable generation during inference. However, effectively training large diffusion policies using reinforcement learning (RL) remains challenging. Existing methods either suffer from unstable training due to directly maximizing value objectives, or face computational issues due to relying on crude Gaussian likelihood approximation, which requires a large amount of sufficiently small denoising steps. In this work, we propose DIPOLE (Dichotomous diffusion Policy improvement), a novel RL algorithm designed for stable and controllable diffusion policy optimization. We begin by revisiting the KL-regularized objective in RL, which offers a desirable weighted regression objective for diffusion policy extraction, but often struggles to balance greediness and stability. We then formulate a greedified policy regularization scheme, which naturally enables decomposing the optimal policy into a pair of stably learned dichotomous policies: one aims at reward maximization, and the other focuses on reward minimization. Under such a design, optimized actions can be generated by linearly combining the scores of dichotomous policies during inference, thereby enabling flexible control over the level of greediness.Evaluations in offline and offline-to-online RL settings on ExORL and OGBench demonstrate the effectiveness of our approach. We also use DIPOLE to train a large vision-language-action (VLA) model for end-to-end autonomous driving (AD) and evaluate it on the large-scale real-world AD benchmark NAVSIM, highlighting its potential for complex real-world applications.

cs.LG↗

arXiv2026

Horizon Adaptive Offline Policy Learning via Value Stitching

Learning accurate value functions plays a decisive role for reinforcement learning (RL) agents to solve long-horizon, complex tasks. Conventional temporal-difference (TD) learning objectives suffer from value-estimation bias that accumulates over the horizon, while extended-horizon modeling methods, such as n-step TD backups and Q-chunking, adopt a rigid, fixed-horizon value-modeling recipe that is often not flexible enough to capture complex value structures in long-horizon, multi-stage tasks. In this paper, we show that enabling value updates with dynamic horizon composition can yield a strong offline policy learning scheme. Our method, Horizon Adaptive Offline Policy Learning via VAlue STitching (VAST), replaces fixed-horizon backups with recursive, horizon-adaptive value composition. Its key ingredient is to couple value optimization with a future state- and horizon-length-conditioned auxiliary value function that is learned through direct data supervision, and a stitching policy that optimally selects the reward-maximizing horizon length and future sub-goal to achieve horizon-adaptive value stitching. This design enables direct estimation and compositional "stitching" of variable-length returns grounded in actionable sub-goal states, providing an accurate and greedily exploitable value-supervision signal for offline policy optimization. Across 50 tasks on OGBench, VAST outperforms fixed-step, extended-horizon methods, and generative-value offline RL baselines, achieving strong performance particularly in high-complexity, long-horizon decision-making tasks.

cs.LG↗

arXiv2026

GameplayQA: A Benchmarking Framework for Decision-Dense POV-Synced Multi-Video Understanding of 3D Virtual Agents

Multimodal LLMs are increasingly deployed as perceptual backbones for autonomous agents in 3D environments, from robotics to virtual worlds. These applications require agents to perceive rapid state changes, attribute actions to the correct entities, and reason about concurrent multi-agent behaviors from a first-person perspective, capabilities that existing benchmarks do not adequately evaluate. We introduce GameplayQA, a framework for evaluating agentic-centric perception and reasoning through video understanding. Specifically, we densely annotate multiplayer 3D gameplay videos at 1.22 labels/second, with time-synced, concurrent captions of states, actions, and events structured around a triadic system of Self, Other Agents, and the World, a natural decomposition for multi-agent environments. From these annotations, we refined 2.4K diagnostic QA pairs organized into three levels of cognitive complexity, accompanied by a structured distractor taxonomy that enables fine-grained analysis of where models hallucinate. Evaluation of frontier MLLMs reveals a substantial gap from human performance, with common failures in temporal and cross-video grounding, agent-role attribution, and handling the decision density of the game. We hope GameplayQA stimulates future research at the intersection of embodied AI, agentic perception, and world modeling.

cs.CL↗

arXiv2025

Personalized Transcranial Electrical Stimulation: A Review of Computational Modeling and Optimization

Objective. Personalized transcranial electrical stimulation (tES) has gained growing attention due to the substantial inter-individual variability in brain anatomy and physiology. While previous reviews have discussed the physiological mechanisms and clinical applications of tES, there remains a critical gap in up-to-date syntheses focused on the computational modeling frameworks that enable individualized stimulation optimization. Approach. This review presents a comprehensive overview of recent advances in computational techniques supporting personalized tES. We systematically examine developments in forward modeling for simulating individualized electric fields, as well as inverse modeling approaches for optimizing stimulation parameters. We critically evaluate progress in head modeling pipelines, optimization algorithms, and the integration of multimodal brain data. Main results. Recent advances have substantially accelerated the construction of subject-specific head conductor models and expanded the landscape of optimization methods, including multi-objective optimization and brain network-informed optimization. These advances allow for dynamic and individualized stimulation planning, moving beyond empirical trial-and-error approaches.Significance. By integrating the latest developments in computational modeling for personalized tES, this review highlights current challenges, emerging opportunities, and future directions for achieving precision neuromodulation in both research and clinical contexts.

q-bio.NC↗

arXiv2025

Contact Map Transfer with Conditional Diffusion Model for Generalizable Dexterous Grasp Generation

Dexterous grasp generation is a fundamental challenge in robotics, requiring both grasp stability and adaptability across diverse objects and tasks. Analytical methods ensure stable grasps but are inefficient and lack task adaptability, while generative approaches improve efficiency and task integration but generalize poorly to unseen objects and tasks due to data limitations. In this paper, we propose a transfer-based framework for dexterous grasp generation, leveraging a conditional diffusion model to transfer high-quality grasps from shape templates to novel objects within the same category. Specifically, we reformulate the grasp transfer problem as the generation of an object contact map, incorporating object shape similarity and task specifications into the diffusion process. To handle complex shape variations, we introduce a dual mapping mechanism, capturing intricate geometric relationship between shape templates and novel objects. Beyond the contact map, we derive two additional object-centric maps, the part map and direction map, to encode finer contact details for more stable grasps. We then develop a cascaded conditional diffusion model framework to jointly transfer these three maps, ensuring their intra-consistency. Finally, we introduce a robust grasp recovery mechanism, identifying reliable contact points and optimizing grasp configurations efficiently. Extensive experiments demonstrate the superiority of our proposed method. Our approach effectively balances grasp quality, generation efficiency, and generalization performance across various tasks. Project homepage: https://cmtdiffusion.github.io/

cs.RO↗

arXiv2025

Towards Robust Zero-Shot Reinforcement Learning

The recent development of zero-shot reinforcement learning (RL) has opened a new avenue for learning pre-trained generalist policies that can adapt to arbitrary new tasks in a zero-shot manner. While the popular Forward-Backward representations (FB) and related methods have shown promise in zero-shot RL, we empirically found that their modeling lacks expressivity and that extrapolation errors caused by out-of-distribution (OOD) actions during offline learning sometimes lead to biased representations, ultimately resulting in suboptimal performance. To address these issues, we propose Behavior-REgularizEd Zero-shot RL with Expressivity enhancement (BREEZE), an upgraded FB-based framework that simultaneously enhances learning stability, policy extraction capability, and representation learning quality. BREEZE introduces behavioral regularization in zero-shot RL policy learning, transforming policy optimization into a stable in-sample learning paradigm. Additionally, BREEZE extracts the policy using a task-conditioned diffusion model, enabling the generation of high-quality and multimodal action distributions in zero-shot RL settings. Moreover, BREEZE employs expressive attention-based architectures for representation modeling to capture the complex relationships between environmental dynamics. Extensive experiments on ExORL and D4RL Kitchen demonstrate that BREEZE achieves the best or near-the-best performance while exhibiting superior robustness compared to prior offline zero-shot RL methods. The official implementation is available at: https://github.com/Whiterrrrr/BREEZE.

cs.LG↗

arXiv2025

Flow Matching-Based Autonomous Driving Planning with Advanced Interactive Behavior Modeling

Modeling interactive driving behaviors in complex scenarios remains a fundamental challenge for autonomous driving planning. Learning-based approaches attempt to address this challenge with advanced generative models, removing the dependency on over-engineered architectures for representation fusion. However, brute-force implementation by simply stacking transformer blocks lacks a dedicated mechanism for modeling interactive behaviors that are common in real driving scenarios. The scarcity of interactive driving data further exacerbates this problem, leaving conventional imitation learning methods ill-equipped to capture high-value interactive behaviors. We propose Flow Planner, which tackles these problems through coordinated innovations in data modeling, model architecture, and learning scheme. Specifically, we first introduce fine-grained trajectory tokenization, which decomposes the trajectory into overlapping segments to decrease the complexity of whole trajectory modeling. With a sophisticatedly designed architecture, we achieve efficient temporal and spatial fusion of planning and scene information, to better capture interactive behaviors. In addition, the framework incorporates flow matching with classifier-free guidance for multi-modal behavior generation, which dynamically reweights agent interactions during inference to maintain coherent response strategies, providing a critical boost for interactive scenario understanding. Experimental results on the large-scale nuPlan dataset and challenging interactive interPlan dataset demonstrate that Flow Planner achieves state-of-the-art performance among learning-based approaches while effectively modeling interactive behaviors in complex driving scenarios.

cs.RO↗

arXiv2025

Diffusion-Based Planning for Autonomous Driving with Flexible Guidance

Achieving human-like driving behaviors in complex open-world environments is a critical challenge in autonomous driving. Contemporary learning-based planning approaches such as imitation learning methods often struggle to balance competing objectives and lack of safety assurance,due to limited adaptability and inadequacy in learning complex multi-modal behaviors commonly exhibited in human planning, not to mention their strong reliance on the fallback strategy with predefined rules. We propose a novel transformer-based Diffusion Planner for closed-loop planning, which can effectively model multi-modal driving behavior and ensure trajectory quality without any rule-based refinement. Our model supports joint modeling of both prediction and planning tasks under the same architecture, enabling cooperative behaviors between vehicles. Moreover, by learning the gradient of the trajectory score function and employing a flexible classifier guidance mechanism, Diffusion Planner effectively achieves safe and adaptable planning behaviors. Evaluations on the large-scale real-world autonomous planning benchmark nuPlan and our newly collected 200-hour delivery-vehicle driving dataset demonstrate that Diffusion Planner achieves state-of-the-art closed-loop performance with robust transferability in diverse driving styles.

cs.RO↗

arXiv2024

Generative AI-driven forecasting of oil production

Forecasting oil production from oilfields with multiple wells is an important problem in petroleum and geothermal energy extraction, as well as energy storage technologies. The accuracy of oil forecasts is a critical determinant of economic projections, hydrocarbon reserves estimation, construction of fluid processing facilities, and energy price fluctuations. Leveraging generative AI techniques, we model time series forecasting of oil and water productions across four multi-well sites spanning four decades. Our goal is to effectively model uncertainties and make precise forecasts to inform decision-making processes at the field scale. We utilize an autoregressive model known as TimeGrad and a variant of a transformer architecture named Informer, tailored specifically for forecasting long sequence time series data. Predictions from both TimeGrad and Informer closely align with the ground truth data. The overall performance of the Informer stands out, demonstrating greater efficiency compared to TimeGrad in forecasting oil production rates across all sites.

cs.LG↗

arXiv2024

Orbital Angular Momentum Active Anti-Jamming in Radio Wireless Communications

Orbital angular momentum (OAM), providing the orthogonality among different OAM modes, has attracted much attention to significantly increase spectrum efficiencies (SEs) and enhance the anti-jamming results of wireless communications. However, the SE of wireless communications is severely degraded under co-frequency and co-mode hostile jamming. Focused on this issue, we propose a novel OAM active anti-jamming scheme to significantly enhance the anti-jamming results of wireless communications under broadband hostile jamming. Specifically, the OAM transmitter with energy detection senses jamming signals to identify which OAM modes are jammed and unjammed. Based on the recognition of OAM modes, useful signals are modulated by reflecting the received co-frequency and co-mode jamming signals with the assistance of a programmable gain amplifier (PGA) to the OAM receiver, thus utilizing both the OAM modes jammed by hostile attacks and the energy of jamming signals. Meanwhile, the unjammed OAM modes allocated with total transmit power are multiplexed for useful signal transmission. Numerical results demonstrate that our proposed OAM active anti-jamming scheme can achieve high OAM mode utilization and significantly increase the SEs.

eess.SP↗

arXiv2024

M3FAS: An Accurate and Robust MultiModal Mobile Face Anti-Spoofing System

Face presentation attacks (FPA), also known as face spoofing, have brought increasing concerns to the public through various malicious applications, such as financial fraud and privacy leakage. Therefore, safeguarding face recognition systems against FPA is of utmost importance. Although existing learning-based face anti-spoofing (FAS) models can achieve outstanding detection performance, they lack generalization capability and suffer significant performance drops in unforeseen environments. Many methodologies seek to use auxiliary modality data (e.g., depth and infrared maps) during the presentation attack detection (PAD) to address this limitation. However, these methods can be limited since (1) they require specific sensors such as depth and infrared cameras for data capture, which are rarely available on commodity mobile devices, and (2) they cannot work properly in practical scenarios when either modality is missing or of poor quality. In this paper, we devise an accurate and robust MultiModal Mobile Face Anti-Spoofing system named M3FAS to overcome the issues above. The primary innovation of this work lies in the following aspects: (1) To achieve robust PAD, our system combines visual and auditory modalities using three commonly available sensors: camera, speaker, and microphone; (2) We design a novel two-branch neural network with three hierarchical feature aggregation modules to perform cross-modal feature fusion; (3). We propose a multi-head training strategy, allowing the model to output predictions from the vision, acoustic, and fusion heads, resulting in a more flexible PAD. Extensive experiments have demonstrated the accuracy, robustness, and flexibility of M3FAS under various challenging experimental settings. The source code and dataset are available at: https://github.com/ChenqiKONG/M3FAS/

cs.MM↗