Searcharxiv⌕ Search

arXiv subjects

Hua Chen

Publications and source records attributed to Hua Chen.

At least 37 records · Page 2Linked to original sources
arXiv2026

Any2Any: Efficient Cross-Embodiment Transfer for Humanoid Whole-Body Tracking

Whole-body tracking (WBT) models have become a key foundation for humanoid robots, enabling them to imitate diverse motions with high fidelity. Training such models from scratch requires large-scale data and computation, making rapid deployment on new humanoid platforms costly. This raises a natural question: Can pretrained WBT models transfer across embodiments with minimal adaptation? To answer this question, we propose Any2Any, a paradigm that efficiently transfers an existing WBT specialist to a new humanoid embodiment with only a small amount of data and compute. Any2Any first performs kinematic alignment between source and target humanoids, aligning their input and output spaces so that the pretrained source policy can be meaningfully reused on the target embodiment.Any2Any then performs dynamics adaptation by applying lightweight parameter-efficient fine-tuning (PEFT) components to selected dynamics-sensitive modules, preserving useful behavioral priors while enabling targeted adaptation to the target robot. Extensive experiments on multiple humanoid platforms and pretrained backbones show that Any2Any substantially accelerates convergence and reduces training cost compared with training from scratch, while achieving competitive or superior tracking performance. Notably, using only 1% of the compute and data required for full training, Any2Any successfully transfers Sonic models pre-trained on Unitree G1 to LimX Oli and LimX Luna. These results suggest that pretrained WBT specialists can be efficiently reused across embodiments, providing a scalable path toward deploying humanoid whole-body control on new robots. More results and videos are available on our project page: https://any2any.top/.

cs.RO↗
arXiv2026

Existence results for nonlinear cone degenerate Laplace equations

This paper concerns a class of non-divergence nonlinear elliptic equations driven by the cone degenerate Laplacian, which is motivated by cone calculus. We establish the existence of viscosity solutions by proving the Alexandrov-Bakelman-Pucci and Hölder estimates. Furthermore, we obtain the existence of weak solutions by proving the equivalence between weak solutions and viscosity solutions.

math.AP↗
arXiv2026

TideGS: Scalable Training of Over One Billion 3D Gaussian Splatting Primitives via Out-of-Core Optimization

Training 3D Gaussian Splatting (3DGS) at billion-primitive scale is fundamentally memory-bound: each Gaussian primitive carries a large attribute vector, and the aggregate parameter table quickly exceeds GPU capacity, limiting prior systems to tens of millions of Gaussians on commodity single-GPU hardware. We observe that 3DGS training is inherently sparse and trajectory-conditioned: each iteration activates only the Gaussians visible from the current camera batch, so GPU memory can serve as a working-set cache rather than a persistent parameter store. Building on this insight, we introduce TideGS, an out-of-core training framework that manages parameters across an SSD-CPU-GPU hierarchy via three synergistic techniques: block-virtualized geometry for SSD-aligned spatial locality, a hierarchical asynchronous pipeline to overlap I/O with computation, and trajectory-adaptive differential streaming that transfers only incremental working-set deltas between iterations. Experiments show that TideGS enables training with over one billion Gaussians on a single 24 GB GPU while achieving the best reconstruction quality among evaluated single-GPU baselines on large-scale scenes, scaling beyond prior out-of-core baselines (e.g., approximately 100M Gaussians) and standard in-memory training (e.g., approximately 11M Gaussians).

cs.CV↗
arXiv2026

CktGen: Automated Analog Circuit Design with Generative Artificial Intelligence

The automatic synthesis of analog circuits presents significant challenges. Most existing approaches formulate the problem as a single-objective optimization task, overlooking that design specifications for a given circuit type vary widely across applications. To address this, we introduce specification-conditioned analog circuit generation, a task that directly generates analog circuits based on target specifications. The motivation is to leverage existing well-designed circuits to improve automation in analog circuit design. Specifically, we propose CktGen, a simple yet effective variational autoencoder that maps discretized specifications and circuits into a joint latent space and reconstructs the circuit from that latent vector. Notably, as a single specification may correspond to multiple valid circuits, naively fusing specification information into the generative model does not capture these one-to-many relationships. To address this, we decouple the encoding of circuits and specifications and align their mapped latent space. Then, we employ contrastive training with a filter mask to maximize differences between encoded circuits and specifications. Furthermore, classifier guidance along with latent feature alignment promotes the clustering of circuits sharing the same specification, avoiding model collapse into trivial one-to-one mappings. By canonicalizing the latent space with respect to specifications, we can search for an optimal circuit that meets valid target specifications. We conduct comprehensive experiments on the open circuit benchmark and introduce metrics to evaluate cross-model consistency. Experimental results demonstrate that CktGen achieves substantial improvements over state-of-the-art methods.

cs.LG↗
arXiv2026

Beyond Pixels: Learning Invariant Rewards for Real-World Robotics From a Few Demonstrations

Designing reward functions that generalize beyond controlled laboratory settings remains a fundamental challenge in reinforcement learning for robotics. In open-world manipulation problems, a single task can appear in numerous variants through different object instances, positions, and camera viewpoints. Recent vision-based reward models tend to memorize specific pixel distributions and fail to generalize beyond their training conditions. To address this, we propose a framework that learns invariant symbolic reward functions from as few as five demonstrations. The insight is to shift from visual feature-fitting to the discovery of behavioral invariants: task-level properties that remain constant across diverse visual instantiations. The framework has two coupled components: a structural reward formulation that encodes task-level strategies and physical constraints while preserving optimal policy invariance, and a hybrid symbolic-numerical procedure that distills these invariants from demonstrations without online interaction. Experiments on eight Meta-World tasks and three Franka manipulation tasks demonstrate that our method achieves stronger process alignment and policy rollout ranking abilities compared to baselines, accelerating downstream policy learning. Three real-world out-of-distribution experiments further show that the same learned reward generalizes zero-shot to position, viewpoint, and object variations, enabling a single reward representation to be reused across diverse task variants in practice.

cs.RO↗
arXiv2026

Compute Only Once: UG-Separation for Efficient Large Recommendation Models

Driven by scaling laws, recommender systems increasingly rely on larger-scale models to capture complex feature interactions and user behaviors, but this trend also leads to prohibitive training and inference costs. While long-sequence models can reuse user-side computation through KV Caching, such reuse is difficult in TokenMixer-based dense feature interaction architectures, where user and group features are deeply entangled and mixed-up across layers. In this work, we present User-Group Separation (UG-Sep), an industrial large-scale framework that enables user-side computation reusable in TokenMixer-based dense interaction models for the first time. UG-Sep explicitly disentangles user-side and item-side information flows within token-mixing layers, ensuring that a subset of tokens preserves purely user-side representations across layers. This design allows the corresponding per-token computations to be reused across multiple samples, significantly reducing redundant inference cost. To compensate for the potential expressive capacity loss induced by masking, we further propose an Information Compensation strategy that adaptively reconstructs suppressed user-item interactions. Moreover, as UG-Sep substantially reduces user-side FLOPs and exposes memory-bound components, we incorporate W8A16 (8-bit weight, 16-bit activation) weight-only quantization to alleviate memory bandwidth bottlenecks and achieve additional acceleration. We conduct extensive offline evaluations and large-scale online A/B experiments at ByteDance to validate the effectiveness of UG-Sep. Results show that UG-Sep reduces inference latency by up to 20% without causing adverse changes to online user experience and commercial metrics on multiple influential business scenarios compared to TokenMixer at ByteDance, including Douyin Feed Recommendation, Hongguo Feed Recommendation, Chuanshanjia Ads, and Qianchuan Ads.

cs.IR↗
arXiv2026

ARM: Advantage Reward Modeling for Long-Horizon Manipulation

Long-horizon robotic manipulation remains challenging for reinforcement learning (RL) because sparse rewards provide limited guidance for credit assignment. Practical policy improvement thus relies on richer intermediate supervision, such as dense progress rewards, which are costly to obtain and ill-suited to non-monotonic behaviors such as backtracking and recovery. To address this, we propose Advantage Reward Modeling (ARM), a framework that shifts from hard-to-quantify absolute progress to estimating relative advantage. We introduce a cost-effective tri-state labeling strategy -- Progressive, Regressive, and Stagnant -- that reduces human cognitive overhead while ensuring high cross-annotator consistency. By training on these intuitive signals, ARM enables automated progress annotation for both complete demonstrations and fragmented DAgger-style data. Integrating ARM into an offline RL pipeline allows for adaptive action-reward reweighting, effectively filtering suboptimal samples. Our approach achieves a 99.4% success rate on a challenging long-horizon towel-folding task, demonstrating improved stability and data efficiency over current VLA baselines with near-zero human intervention during policy training.

cs.RO↗
arXiv2026

Theoretical and observational constraints on early dark energy in $F(R)$ gravity

This work examines an early dark energy (EDE) scenario in the context of $F(R)$ gravity. EDE is introduced to alleviate the Hubble tension by temporarily injecting approximately $10\%$ of the energy fraction around the matter-radiation equality epoch ($z \approx 10^{3}$--$10^{4}$). Building on several benchmark models, we focus on the potential-driven EDE scenario and investigate the conditions required within $F(R)$ gravity. We first introduce a dimensionless quantity to analytically visualize the evolution of the density ratio between EDE and other matter components. Considering several examples, we demonstrate that the desired EDE can indeed be realized in $F(R)$ gravity. However, stringent constraints arising from violations of the equivalence principle could exclude the allowed parameter space. Our result provides a generic constraint on the potential-driven EDE in $F(R)$ gravity at the background level. This work also concludes that nonperturbative effects or nontrivial mechanisms are indispensable for studying EDE in $F(R)$ gravity while maintaining compatibility with local tests of gravity.

gr-qc↗
arXiv2026

3D XY Universality and Nonlinear magnetic susceptibility in a kagome ice compound

Kagome spin ice is an intriguing class of spin systems constituted by in-plane Ising spins with ferromagnetic interaction residing on the kagome lattice, theoretically predicted to host a plethora of magnetic transitions and excitations. In particular, different variants of kagome spin ice models can exhibit different sequences of symmetry breaking upon cooling from the paramagnetic to the fully ordered ground state. Recently, it has been demonstrated that the frustrated intermetallic HoAgGe stands as a faithful solid-state realization of kagome spin ice. Here we use single crystal neutron diffuse scattering to map the spin ordering of HoAgGe at various temperatures more accurately and surprisingly find that the ordering sequence appears to be different from previously known scenarios: From the paramagnetic state, the system first enters a partially ordered state with fluctuating magnetic charges, in contrast to a charge-ordered paramagnetic phase before reaching the fully ordered state. Through state-of-the-art Monte Carlo simulations and scaling analyses using a quasi-2D model for the distorted Kagome spin ice in HoAgGe, we elucidate a single three-dimensional (3D) XY phase transition into the ground state with broken time-reversal symmetry (TRS). However, the 3D XY transition has a long crossover tail before the fluctuating magnetic charges fully order. More interestingly, we find both experimentally and theoretically that the TRS breaking phase of HoAgGe features an unusual, hysteretic response: In spite of their vanishing magnetization, the two time-reversal partners are distinguished and selected by a nonlinear magnetic susceptibility tied to the kagome ice rule. Our discovery not only unveils a new symmetry breaking hierarchy of kagome spin ice, but also demonstrates the potential of TRS-breaking frustrated spin systems for information technology applications.

cond-mat.str-el↗
arXiv2026

First-principles theory of spin magnetic multipole moments in antiferromagnets

Antiferromagnets with vanishing net magnetization are naturally expected to host higher-order magnetic multipole moments. Understanding and utilizing the multipole degrees of freedom are imperative for novel conceptual designs and applications unique to antiferromagnets. However, a universal, quantitative definition of magnetic multipole moments of antiferromagnetic materials is currently lacking. In this work we provide a unified description of arbitrary-order spin magnetic multipole moments (SM$^3$) of antiferromagnets by introducing a nonlocal spin density in macroscopic Maxwell equations. The formalism makes it transparent how SM$^3$ calculated for translationally invariant bulk systems corresponds to experimental observables when translation symmetry is broken. Through the nonlocal spin density calculated from first principles, we propose a robust scheme to extract arbitrary-order SM$^3$ through symmetry-constrained fitting at long wavelengths. Using this approach, we have calculated SM$^3$ of a few representative antiferromagnets, including $α$-$\rm Fe_2O_3$, Mn$_3$Sn, and Mn$_3$NiN. Moreover, we clarify the role of spin-orbit coupling (SOC) in SM$^3$, especially in the weak SOC limit where clean predictions can be made based on symmetry principles. Our work paves the way for systematically investigating multipolar order parameters of unconventional magnetic materials.

cond-mat.mtrl-sci↗
arXiv2026

Where-to-Learn: Analytical Policy Gradient Directed Exploration for On-Policy Robotic Reinforcement Learning

On-policy reinforcement learning (RL) algorithms have demonstrated great potential in robotic control, where effective exploration is crucial for efficient and high-quality policy learning. However, how to encourage the agent to explore the better trajectories efficiently remains a challenge. Most existing methods incentivize exploration by maximizing the policy entropy or encouraging novel state visiting regardless of the potential state value. We propose a new form of directed exploration that uses analytical policy gradients from a differentiable dynamics model to inject task-aware, physics-guided guidance, thereby steering the agent towards high-reward regions for accelerated and more effective policy learning.

cs.RO↗
arXiv2026

Near-Field NLOS Localization via Position-Unknown HRIS:From Self-Localization to Target Positioning

Current reconfigurable intelligent surface (RIS)-aided near-field (NF) localization methods assume the RIS position is known a priori, and it has limited their practical applicability. This paper applies a hybrid RIS (HRIS) at an unknown position to locate non-line-of-sight (NLOS) NF targets. To this end, we first propose a two-stage gridless localization framework for achieving HRIS self-localization, and then determine the positions of the NF targets. In the first stage, we use the NF Fresnel approximation to convert the signal model into a virtual far-field model through delay-based cross-correlation of centrally symmetric HRIS elements. Such a conversion will naturally extend the aperture of the virtual array. A single-snapshot decoupled atomic norm minimization (DANM) algorithm is then proposed to locate an NF target relative to the HRIS, which includes a two-dimensional (2-D) direction of arrival (DOA) estimation with automatic pairing, the multiple signal classification (MUSIC) method for range estimation, and a total least squares (TLS) method to eliminate the Fresnel approximation error. In the second stage, we leverage the unique capability of HRIS in simultaneous sensing and reflection to estimate the HRIS-to-base station (BS) direction vectors using atomic norm minimization (ANM), and derive the three-dimensional (3-D) HRIS position with two BSs via the least squares (LS)-based geometric triangulation. Furthermore, we propose a semidefinite relaxation (SDR)-based HRIS phase optimization method to enhance the received signal power at the BSs, thereby improving the HRIS localization accuracy, which, in turn, enhances NF target positionings. The Cramer-Rao bound (CRB) for the NF target parameters and the position error bound (PEB) for the HRIS coordinates are derived as performance benchmarks.

eess.SP↗
arXiv2026

Near-Field Multiuser Beam Training for XL-MIMO: An End-to-End Interference-Aware Approach with Pilot Limitations

Near-field propagation in extremely large-scale MIMO (XL-MIMO) enlarges the beam training (BT) search space by introducing an additional range dimension, which makes conventional codebook-based beam sweeping prohibitively expensive under limited pilot resources, especially for multiuser sub-connected hybrid architectures. This letter proposes a deep-learning-based interference-aware multiuser BT framework (DL-IABT) that directly predicts analog beam indices from a small number of uplink sensing measurements. By exploiting a subarray-level approximation, a far-field codebook is adopted to represent each subarray response with negligible mismatch. To enable end-to-end (E2E) learning, we derive a variant-MSE surrogate loss by eliminating the digital precoder through a closed-form MMSE solution from KKT conditions, which implicitly accounts for multiuser interference (MUI). The proposed network integrates a complex-valued sensing front-end, a shared complex-valued encoder, a Transformer-based multiuser predictor, and a scalable Gumbel--Softmax beam selection head. Simulation results show that DL-IABT achieves near-optimal sum-rate performance while providing markedly higher effective throughput under pilot overhead constraints.

eess.SP↗
arXiv2026

PvP: Data-Efficient Humanoid Robot Learning with Proprioceptive-Privileged Contrastive Representations

Achieving efficient and robust whole-body control (WBC) is essential for enabling humanoid robots to perform complex tasks in dynamic environments. Despite the success of reinforcement learning (RL) in this domain, its sample inefficiency remains a significant challenge due to the intricate dynamics and partial observability of humanoid robots. To address this limitation, we propose PvP, a Proprioceptive-Privileged contrastive learning framework that leverages the intrinsic complementarity between proprioceptive and privileged states. PvP learns compact and task-relevant latent representations without requiring hand-crafted data augmentations, enabling faster and more stable policy learning. To support systematic evaluation, we develop SRL4Humanoid, the first unified and modular framework that provides high-quality implementations of representative state representation learning (SRL) methods for humanoid robot learning. Extensive experiments on the LimX Oli robot across velocity tracking and motion imitation tasks demonstrate that PvP significantly improves sample efficiency and final performance compared to baseline SRL methods. Our study further provides practical insights into integrating SRL with RL for humanoid WBC, offering valuable guidance for data-efficient humanoid robot learning.

cs.RO↗
arXiv2026

Diffusion Policy through Conditional Proximal Policy Optimization

Reinforcement learning (RL) has been extensively employed in a wide range of decision-making problems, such as games and robotics. Recently, diffusion policies have shown strong potential in modeling multi-modal behaviors, enabling more diverse and flexible action generation compared to the conventional Gaussian policy. Despite various attempts to combine RL with diffusion, a key challenge is the difficulty of computing action log-likelihood under the diffusion model. This greatly hinders the direct application of diffusion policies in on-policy reinforcement learning. Most existing methods calculate or approximate the log-likelihood through the entire denoising process in the diffusion model, which can be memory- and computationally inefficient. To overcome this challenge, we propose a novel and efficient method to train a diffusion policy in an on-policy setting that requires only evaluating a simple Gaussian probability. This is achieved by aligning the policy iteration with the diffusion process, which is a distinct paradigm compared to previous work. Moreover, our formulation can naturally handle entropy regularization, which is often difficult to incorporate into diffusion policies. Experiments demonstrate that the proposed method produces multimodal policy behaviors and achieves superior performance on a variety of benchmark tasks in both IsaacLab and MuJoCo Playground.

cs.LG↗
arXiv2026

BFA++: Hierarchical Best-Feature-Aware Token Prune for Multi-View Vision Language Action Model

Vision-Language-Action (VLA) models have achieved significant breakthroughs by leveraging Large Vision Language Models (VLMs) to jointly interpret instructions and visual inputs. However, the substantial increase in visual tokens, particularly from multi-view inputs, poses serious challenges to real-time robotic manipulation. Existing acceleration techniques for VLMs, such as token pruning, often result in degraded performance when directly applied to VLA models, as they overlook the relationships between different views and fail to account for the dynamic and task-specific characteristics of robotic operation. To address this, we propose BFA++, a dynamic token pruning framework designed specifically for VLA models. BFA++ introduces a hierarchical pruning strategy guided by two-level importance predictors: an intra-view predictor highlights task-relevant regions within each image to suppress spatial noise, while an inter-view predictor identifies critical camera views throughout different manipulation phases to reduce cross-view redundancy. This design enables efficient token selection while preserving essential visual cues, resulting in improved computational efficiency and higher manipulation success rates. Evaluations on the RoboTwin benchmark and real-world robotic tasks demonstrate that BFA++ consistently outperforms existing methods. BFA++ improves the success rate by about 10% on both the π0 and RDT models, achieving speedup of 1.8X and 1.5X, respectively. Our results highlight that context-sensitive and task-aware token pruning serves as a more effective strategy than full visual processing, enabling faster inference and improved manipulation accuracy in real-world robotic systems.

cs.RO↗
arXiv2026

A Survey of Behavior Foundation Model: Next-Generation Whole-Body Control System of Humanoid Robots

Humanoid robots are drawing significant attention as versatile platforms for complex motor control, human-robot interaction, and general-purpose physical intelligence. However, achieving efficient whole-body control (WBC) in humanoids remains a fundamental challenge due to sophisticated dynamics, underactuation, and diverse task requirements. While learning-based controllers have shown promise for complex tasks, their reliance on labor-intensive and costly retraining for new scenarios limits real-world applicability. To address these limitations, behavior(al) foundation models (BFMs) have emerged as a new paradigm that leverages large-scale pre-training to learn reusable primitive skills and broad behavioral priors, enabling zero-shot or rapid adaptation to a wide range of downstream tasks. In this paper, we present a comprehensive overview of BFMs for humanoid WBC, tracing their development across diverse pre-training pipelines. Furthermore, we discuss real-world applications, current limitations, urgent challenges, and future opportunities, positioning BFMs as a key approach toward scalable and general-purpose humanoid intelligence. Finally, we provide a curated and regularly updated collection of BFM papers and projects to facilitate more subsequent research, which is available at https://github.com/yuanmingqi/awesome-bfm-papers.

cs.RO↗
arXiv2026

Beyond $λ/2$: Can Arbitrary EMVS Arrays Achieve Unambiguous NLOS Localization?

Conventional radar array design mandates interelement spacing not exceeding half a wavelength ($λ/2$) to avoid spatial ambiguity, fundamentally limiting array aperture and angular resolution. This paper addresses the fundamental question: Can arbitrary electromagnetic vector sensor (EMVS) arrays achieve unambiguous reconfigurable intelligent surface (RIS)-aided localization when element spacing exceeds $λ/2$? We provide an affirmative answer by exploiting the multi-component structure of EMVS measurements and developing a synergistic estimation and optimization framework for non-line-of-sight (NLOS) bistatic multiple input multiple output (MIMO) radar. A third-order parallel factor (PARAFAC) model is constructed from EMVS observations, enabling natural separation of spatial, polarimetric, and propagation effects via the trilinear alternating least squares (TALS) algorithm. A novel phase-disambiguation procedure leverages rotational invariance across the six electromagnetic components of EMVSs to resolve $2π$ phase wrapping in arbitrary array geometries, allowing unambiguous joint estimation of two-dimensional (2-D) direction of departure (DOD), two-dimensional direction of arrival (DOA), and polarization parameters with automatic pairing. To support localization in NLOS environments and enhance estimation robustness, a reconfigurable intelligent surface (RIS) is incorporated and its phase shifts are optimized via semidefinite programming (SDP) relaxation to maximize received signal power, improving signal-to-noise ratio (SNR) and further suppressing spatial ambiguities through iterative refinement.

eess.SP↗