arXiv subjects

Jiahao Wu

Publications and source records attributed to Jiahao Wu.

At least 19 recordsLinked to original sources

arXiv2026

SEED-UMI: Sharing the Exoskeleton between human and robot for onE-to-one Dexterous demonstration

Imitation learning for dexterous hands is bottlenecked by the difficulty of collecting contact-rich demonstrations that transfer faithfully to the robot. Prior wearable-exoskeleton systems record only on the human side and retarget via open-loop mappings calibrated in free space, which degrade under contact. We present SEED-UMI, a framework in which both the human and the robot wear the same exoskeleton: joint encoders become a physically shared measurement, and wrist cameras mounted to the exoskeleton observe the same outer mechanism during both human data collection and robot policy rollouts. This turns retargeting into paired cross-embodiment supervision and lets policies train directly on raw exoskeleton-centric wrist images, without segmentation or inpainting. On five contact-rich tasks, SEED-UMI achieves 3.0x greater data collection efficiency than exoskeleton-based teleoperation and a 70.0% average rollout success rate.

cs.RO↗

arXiv2026

Act More, Decide Less: Skill-Guided Adaptive Action Chunking for Long-Horizon LLM Agents

Large language model (LLM) agents for long-horizon interactive tasks typically follow a ReAct-style protocol, issuing one primitive action per LLM round. While this enables frequent replanning, it is inefficient for long-horizon tasks where many rounds are spent on routine action sequences. A natural alternative is to let the agent emit variable-length action chunks. However, naively training such policies with standard reinforcement learning fails: the agent either collapses to single-action behavior or over-commits to excessively long sequences. Both failures share a common root cause: the inability to learn chunk boundaries. We propose SPACE, which addresses this challenge by distilling chunk-boundary supervision from trajectory-induced programmatic skills. We induce two-level programmatic skills from successful trajectories, where subskill boundaries serve as direct chunk-boundary supervision. This temporal structure is then distilled into a primitive-chunk policy via hybrid on-/off-policy optimization with chunk-aware credit assignment. Experiments on ALFWorld and ScienceWorld show that SPACE improves success rates by 7.0%-31.3% over the strongest baseline in each setting while reducing average LLM decision rounds by up to 78.9%.

cs.LG↗

arXiv2026

TrajMind: Chaining Role-Specialized LoRAs for Fast-and-Slow Collective Trajectory Anomaly Diagnosis

Diagnosing collective anomalies from urban trajectories is increasingly important for traffic governance, as it reveals what happened, who was involved, and where and when the event occurred. Existing detectors efficiently produce scores or labels, whereas vision--language pipelines provide richer semantics; neither couples verifiable diagnosis with low-latency monitoring. The central challenge is to recognize collective patterns and recover exact event details from the source trajectories without running the full diagnostic pipeline for every monitored window. We therefore separate always-on screening from on-demand diagnosis: screening raises alerts, while diagnosis releases only source-verified what--who--where--when records. We present TrajMind, a fast-and-slow framework that switches three role-specialized LoRA adapters over one frozen vision--language backbone. Its slow path, \textit{TrajMind$_{\text{slow}}$}, chains canvas-based typing, type-conditioned localization over serialized trajectories, and executable verification, yielding structured, evidence-backed diagnoses. Additionally, the fast path, \textit{TrajMind$_{\text{fast}}$}, screens each window in a single text-only pass, delivering efficient structured alerts. Extensive experiments show that, TrajMind$_{\mathrm{slow}}$ outperforms the strongest baselines by at least $15.3$ percentage points in anomaly typing and $13.8$ percentage points in localization. These gains persist under cross-city transfer, and TrajMind$_{\mathrm{fast}}$ reduces latency by $41.1\%$ and maintains binary balanced accuracy of at least $93.5\%$. Together, TrajMind delivers accurate, evidence-backed diagnoses across cities and efficient front-line monitoring.

cs.LG↗

arXiv2026

ATGS: Anchored Temporal Gaussian Splatting for Long Volumetric Video Representation

Volumetric video enables immersive free viewpoint rendering of dynamic real world scenes, yet existing methods struggle with long sequences and complex motions, often leading to temporal instability and visual artifacts. To address these challenges, we propose \ourname, a Gaussian splatting based framework for volumetric video reconstruction. Our key insight is that explicitly tracking long term complex motion with individual Gaussian primitives is inherently unstable. Instead, we organize Gaussians around time conditioned anchors that localize their spatial and temporal support, thereby reducing long range motion complexity. We further introduce a temporal windowing strategy to activate only anchors relevant to the queried time, which improves scalability and temporal coherence. In addition, to ensure spatial and temporal stability, we design a compact set of multi level anchor features that encode global features, local spatial features, and local temporal features, jointly constraining Gaussian generation. Extensive experiments demonstrate that \ourname \ consistently outperforms prior methods on long sequence volumetric videos with complex motions. Project page: https://github.com/WuJH2001/ATGS.

cs.CV↗

arXiv2026

PRISM: Precision and contact-rich Real-world Industrial Skill dataset with Multimodal sensing

Recent progress in robotic learning has been fueled by large-scale datasets collected in everyday environments. However, most existing datasets emphasize short-horizon, low-contact tasks such as pick-and-place, and therefore do not capture the precision control, force/torque or tactile regulation, and multimodal feedback required for industrial assembly. To address this gap, we introduce PRISM, a large-scale multimodal dataset for contact-rich industrial operations. The dataset spans more than 25 manipulation tasks (e.g., electronic components plug/unplug, conveyor-based sorting) and covers diverse mechanical constraints. PRISM includes more than 5,000 trajectories totaling 45 hours of teleoperated demonstrations, recorded using synchronized multi-view RGB-D, force/torque, tactile, and robot-state measurements. In contrast to datasets collected in household or laboratory settings, PRISM provides a realistic benchmark for multimodal perception and control under high-precision industrial constraints, and serves as a foundation for contact-rich, generalizable manipulation in real-world manufacturing environments. The dataset is open-sourced at: https://tengbo-yu.github.io/PRISM/

cs.RO↗

arXiv2026

TurboRetry: Mitigating Large-Scale QUIC Handshake Floods with Off-the-Shelf DPU Offloading

The modern transport protocol QUIC is designed to enhance network performance and security, but it remains vulnerable to handshake flooding attacks. Such attacks exhaust CPU resources by forcing the server to perform expensive cryptographic operations via a large number of handshaking requests. QUIC provides a built-in defense mechanism, the Retry mechanism, to mitigate these attacks. However, our experiments reveal that it can still become a performance bottleneck under large-scale QUIC handshake floods due to substantial computational overhead. In this paper, we design and implement TurboRetry, a split design, that offloads the Retry mechanism onto DPUs to efficiently mitigate QUIC handshake floods. TurboRetry partitions the tasks of the Retry into two categories, and then assigns them to the DPUs and the host, respectively. To preserve QUIC semantics and reduce the coordination overhead, TurboRetry designs an extended Retry token format and an efficient cooperation scheme. In addition, TurboRetry offloads the connection authorization task to the on-path DPA to further improve both performance and security. Our evaluation shows that TurboRetry outperforms the host-side implementation by a wide margin, improving throughput by 10-20$\times$.

cs.CR↗

arXiv2026

Cooperative Coevolution for Resource-Constrained Agentic LLM Post-Training

Tool-using large language model (LLM) agents produce long, multi-turn trajectories, making gradient-based post-training memory-intensive. Evolution strategies (ES) enable memory-efficient full-parameter post-training without backpropagation and can eventually match the performance of gradient-based reinforcement learning (RL). However, resource-constrained settings typically offer only a few GPUs, so the high GPU-hour requirements of ES translate into prohibitively long training times. To address this, we introduce Cooperative Parameter-subspace Evolution Strategy (CoPES), a cooperative coevolutionary method that decomposes the full parameter space into lower-dimensional subspaces and searches over them cooperatively to improve optimization efficiency. We post-train a Qwen3.5-4B tool-using agent for the math task and evaluate it on five benchmarks of varying difficulty. Under the GPU-hour budget of full-parameter GRPO's best validation checkpoint, CoPES recovers 92% of GRPO's validation-accuracy gain, versus 67% for standard ES, while its theoretical GPU memory requirement is less than one-eighth that of full-parameter GRPO. It consistently outperforms standard ES and LoRA-based GRPO on all evaluated pass@k metrics across the five benchmarks. Additional experiments further show the advantage of CoPES on the question-answering task. These results demonstrate an improved trade-off between memory requirements and training time for agentic LLM post-training under resource constraints. The code is open-sourced in https://github.com/MetaronWang/CoPES

cs.AI↗

arXiv2026

JPPD: Joint Prediction_Planning Diffusion with Differentiable Safety Guidance for Dynamic Obstacle Avoidance in Intelligent Transportation Systems

Shared-space transportation operation requires low-speed autonomous platforms to navigate safely and efficiently among pedestrians, service robots, micromobility users, carts, and other road users. Most existing systems decompose this problem into trajectory prediction followed by motion planning, which creates one-way information flow: predicted participant futures influence the robot plan, but the selected robot plan cannot influence the predicted multi-agent evolution. This paper presents a joint prediction-planning diffusion framework that treats participant prediction and robot planning as a single conditional trajectory generation problem, where the model samples the future robot trajectory and all participant trajectories from one coupled distribution using a causal Transformer with cross-trajectory attention. To replace heuristic repulsive post-processing, the framework introduces differentiable safety potential guidance, a time-varying occupancy-probability potential whose gradient directly guides the joint sampler, and conditional flow matching is used to reduce inference steps while preserving multimodal trajectory diversity. The evaluation emphasizes shared-space operational effects, including near misses, blockage time, induced participant deviation, hard-braking events, and embedded latency, rather than treating average displacement error and final displacement error as the main result. Experiments in scenario-grounded simulation, naturalistic pedestrian replay, Isaac Sim validation, and ROS/Orin deployment show that joint sampling improves tail safety and runtime efficiency over a separated prediction-then-planning baseline.

cs.RO↗

arXiv2026

Policy and World Modeling Co-Training for Language Agents

Reinforcement learning (RL) improves large language model (LLM) agents by teaching them which actions lead to high rewards, but provides little supervision on what those actions do to the environment. World modeling (WM) can fill this gap, yet existing approaches often require separate simulators, extra training stages, or additional inference-time computation. We observe that on-policy RL rollouts already contain the needed signal: each transition pairs an action with its resulting next observation. Based on this observation, we propose PaW, a Policy and World modeling co-training framework that adds auxiliary WM supervision to the same policy during RL, without changing the inference paradigm. To make auxiliary WM supervision informative and stable, PaW introduces three components: action-entropy-based WM data selection, noise-tolerant WM loss, and reward-adaptive loss balancing. Experiments on three agentic task benchmarks show consistent improvements over strong RL baselines across models and RL algorithms. These results suggest that standard RL rollouts are a practical source of WM supervision for language-agent training.

cs.LG↗

arXiv2026

A Mid-Infrared Platform Based on Strontium Tweezer Arrays

Subwavelength atomic tweezer arrays, in which atoms can be positioned at distances smaller than their emission wavelength, have been proposed as a versatile platform to study collective emission phenomena, such as superradiance and subradiance. Experimentally, the realization of such arrays has been a challenge as typical emission wavelengths in the visible or near-infrared are short compared to typical tweezer spacings in the micrometer range. Here, we use $^{88}$Sr atoms in optical tweezer arrays to access a mid-infrared transition at 2,923 nm ($5s5p\:^{3}P_{2} \rightarrow\, 5s4d\:^{3}D_{3}$). We identify a magic trapping wavelength at 597.14(3) nm and demonstrate single-atom preparation and imaging with high fidelity. In addition, using 2,923 nm light, we demonstrate resolved-sideband cooling of tweezer-trapped strontium. Beyond enabling studies of collective emission phenomena in flexible arrangements of atoms, our platform opens novel opportunities for dipolar many-body physics and enhanced control over Rydberg dynamics and the strontium fine-structure qubit.

physics.atom-ph↗

arXiv2026

Neural QAOA$^{2}$: Differentiable Joint Graph Partitioning and Parameter Initialization for Quantum Combinatorial Optimization

The quantum approximate optimization algorithm (QAOA) holds promise for combinatorial optimization but is constrained by limited qubits. While divide-and-conquer frameworks like QAOA$^{2}$ address scalability by partitioning graphs into subgraphs, existing methods suffer from two fundamental limitations: i) misalignment between heuristic partitioning metrics and quantum optimization goals, and ii) topology-blind parameter initialization that leads to optimization cold starts. To bridge these gaps, we propose Neural QAOA$^{2}$, an end-to-end differentiable framework that jointly generates graph partitions and initial parameters. By integrating a generative evaluative network (GEN), our method utilizes a differentiable quantum evaluator as a high-fidelity performance surrogate to provide direct gradient guidance, enabling the joint generator to learn the intrinsic mapping from graph topology to high-quality partition and parameter configurations. Extensive experiments on 183 QUBO, Ising, and MaxCut instances (21 to 1000 variables) demonstrate that our gradient-driven approach broadly outperforms heuristic baselines, ranking first on 101 instances. It exhibits zero-shot generalization across out-of-distribution graph topologies and scales.

quant-ph↗

arXiv2026

Research on mode transition of micro-newton-level cusped field Hall thruster

The micro-newton cusped field Hall thruster is an electric propulsion device that employs microwave-assisted ionization control. It serves as an actuator in drag-free control systems, ensuring control accuracy and stability by providing continuously adjustable thrust over a wide range. However, a mode transition occurring during the regulation process can lead to a sudden change in anode current, degrading control precision and stability. Therefore, it is necessary to investigate the underlying patterns of mode transition. This study examines the variations in internal plasma parameters and discharge characteristics of the thruster before and after microwave mode transition, primarily through probe diagnostics.Experimental results indicate that before the mode transition, the plasma luminous region is primarily concentrated within the electron cyclotron resonance (ECR) area, approximately 1-3 mm upstream of the anode. After the transition, the luminous region moves further upstream, and the plasma density near the anode exceeds the cutoff density, dropping sharply along the axial direction. The fundamental cause of the change in electron heating mechanism is the alteration in the propagation characteristics of fundamental waves due to this plasma density variation.When the plasma density rises to the cutoff density, the R wave and O wave, which drive ionization, are rapidly attenuated or reflected. At this point, the R-wave cannot reach the resonance layer, causing the dominant ECR ionization to become ineffective. The ionization mechanism shifts from being dominated by the R wave and O wave to being dominated primarily by the O wave. Consequently, the electron heating mechanism transitions from volume heating to surface wave heating......

physics.plasm-ph↗

arXiv2026

ClipGStream: Clip-Stream Gaussian Splatting for Any Length and Any Motion Multi-View Dynamic Scene Reconstruction

Dynamic 3D scene reconstruction is essential for immersive media such as VR, MR, and XR, yet remains challenging for long multi-view sequences with large-scale motion. Existing dynamic Gaussian approaches are either Frame-Stream, offering scalability but poor temporal stability, or Clip, achieving local consistency at the cost of high memory and limited sequence length. We propose ClipGStream, a hybrid reconstruction framework that performs stream optimization at the clip level rather than the frame level. The sequence is divided into short clips, where dynamic motion is modeled using clip-independent spatio-temporal fields and residual anchor compensation to capture local variations efficiently, while inter-clip inherited anchors and decoders maintain structural consistency across clips. This Clip-Stream design enables scalable, flicker-free reconstruction of long dynamic videos with high temporal coherence and reduced memory overhead. Extensive experiments demonstrate that ClipGStream achieves state-of-the-art reconstruction quality and efficiency. The project page is available at: https://liangjie1999.github.io/ClipGStreamWeb/

cs.CV↗

arXiv2026

Train at Moving Edge: Online-Verified Prompt Selection for Efficient RL Training of Large Reasoning Model

Reinforcement learning (RL) has become essential for post-training large language models (LLMs) in reasoning tasks. While scaling rollouts can stabilize training and enhance performance, the computational overhead is a critical issue. In algorithms like GRPO, multiple rollouts per prompt incur prohibitive costs, as a large portion of prompts provide negligible gradients and are thus of low utility. To address this problem, we investigate how to select high-utility prompts before the rollout phase. Our experimental analysis reveals that sample utility is non-uniform and evolving: the strongest learning signals concentrate at the ``learning edge", the intersection of intermediate difficulty and high uncertainty, which shifts as training proceeds. Motivated by this, we propose HIVE (History-Informed and online-VErified prompt selection), a dual-stage framework for data-efficient RL. HIVE utilizes historical reward trajectories for coarse selection and employs prompt entropy as a real-time proxy to prune instances with stale utility. By evaluating HIVE across multiple math reasoning benchmarks and models, we show that HIVE yields significant rollout efficiency without compromising performance.

cs.LG↗

arXiv2026

Intrinsic Geometry-Appearance Consistency Optimization for Sparse-View Gaussian Splatting

3D human reconstruction from a single image is a challenging problem and has been exclusively studied in the literature. Recently, some methods have resorted to diffusion models for guidance, optimizing a 3D representation via Score Distillation Sampling(SDS) or generating a back-view image for facilitating reconstruction. However, these methods tend to produce unsatisfactory artifacts (\textit{e.g.} flattened human structure or over-smoothing results caused by inconsistent priors from multiple views) and struggle with real-world generalization in the wild. In this work, we present \emph{MVD-HuGaS}, enabling free-view 3D human rendering from a single image via a multi-view human diffusion model. We first generate multi-view images from the single reference image with an enhanced multi-view diffusion model, which is well fine-tuned on high-quality 3D human datasets to incorporate 3D geometry priors and human structure priors. To infer accurate camera poses from the sparse generated multi-view images for reconstruction, an alignment module is introduced to facilitate joint optimization of 3D Gaussians and camera poses. Furthermore, we propose a depth-based Facial Distortion Mitigation module to refine the generated facial regions, thereby improving the overall fidelity of the reconstruction. Finally, leveraging the refined multi-view images, along with their accurate camera poses, MVD-HuGaS optimizes the 3D Gaussians of the target human for high-fidelity free-view renderings. Extensive experiments on Thuman2.0 and 2K2K datasets show that the proposed MVD-HuGaS achieves state-of-the-art performance on single-view 3D human rendering.

cs.CV↗

arXiv2026

Principled Synthetic Data Enables the First Scaling Laws for LLMs in Recommendation

Large Language Models (LLMs) represent a promising frontier for recommender systems, yet their development has been impeded by the absence of predictable scaling laws, which are crucial for guiding research and optimizing resource allocation. We hypothesize that this may be attributed to the inherent noise, bias, and incompleteness of raw user interaction data in prior continual pre-training (CPT) efforts. This paper introduces a novel, layered framework for generating high-quality synthetic data that circumvents such issues by creating a curated, pedagogical curriculum for the LLM. We provide powerful, direct evidence for the utility of our curriculum by showing that standard sequential models trained on our principled synthetic data significantly outperform ($+130\%$ on recall@100 for SasRec) models trained on real data in downstream ranking tasks, demonstrating its superiority for learning generalizable user preference patterns. Building on this, we empirically demonstrate, for the first time, robust power-law scaling for an LLM that is continually pre-trained on our high-quality, recommendation-specific data. Our experiments reveal consistent and predictable perplexity reduction across multiple synthetic data modalities. These findings establish a foundational methodology for reliable scaling LLM capabilities in the recommendation domain, thereby shifting the research focus from mitigating data deficiencies to leveraging high-quality, structured information.

cs.IR↗

arXiv2026

PEAR: Pixel-aligned Expressive humAn mesh Recovery

Reconstructing detailed 3D human meshes from a single in-the-wild image remains a fundamental challenge in computer vision. Existing SMPLX-based methods often suffer from slow inference, produce only coarse body poses, and exhibit misalignments or unnatural artifacts in fine-grained regions such as the face and hands. These issues make current approaches difficult to apply to downstream tasks. To address these challenges, we propose PEAR-a fast and robust framework for pixel-aligned expressive human mesh recovery. PEAR explicitly tackles three major limitations of existing methods: slow inference, inaccurate localization of fine-grained human pose details, and insufficient facial expression capture. Specifically, to enable real-time SMPLX parameter inference, we depart from prior designs that rely on high resolution inputs or multi-branch architectures. Instead, we adopt a clean and unified ViT-based model capable of recovering coarse 3D human geometry. To compensate for the loss of fine-grained details caused by this simplified architecture, we introduce pixel-level supervision to optimize the geometry, significantly improving the reconstruction accuracy of fine-grained human details. To make this approach practical, we further propose a modular data annotation strategy that enriches the training data and enhances the robustness of the model. Overall, PEAR is a preprocessing-free framework that can simultaneously infer EHM-s (SMPLX and scaled-FLAME) parameters at over 100 FPS. Extensive experiments on multiple benchmark datasets demonstrate that our method achieves substantial improvements in pose estimation accuracy compared to previous SMPLX-based approaches. Project page: https://wujh2001.github.io/PEAR

cs.CV↗

arXiv2026

MANGO:Natural Multi-speaker 3D Talking Head Generation via 2D-Lifted Enhancement

Current audio-driven 3D head generation methods mainly focus on single-speaker scenarios, lacking natural, bidirectional listen-and-speak interaction. Achieving seamless conversational behavior, where speaking and listening states transition fluidly remains a key challenge. Existing 3D conversational avatar approaches rely on error-prone pseudo-3D labels that fail to capture fine-grained facial dynamics. To address these limitations, we introduce a novel two-stage framework MANGO, which leveraging pure image-level supervision by alternately training to mitigate the noise introduced by pseudo-3D labels, thereby achieving better alignment with real-world conversational behaviors. Specifically, in the first stage, a diffusion-based transformer with a dual-audio interaction module models natural 3D motion from multi-speaker audio. In the second stage, we use a fast 3D Gaussian Renderer to generate high-fidelity images and provide 2D-level photometric supervision for the 3D motions through alternate training. Additionally, we introduce MANGO-Dialog, a high-quality dataset with over 50 hours of aligned 2D-3D conversational data across 500+ identities. Extensive experiments demonstrate that our method achieves exceptional accuracy and realism in modeling two-person 3D dialogue motion, significantly advancing the fidelity and controllability of audio-driven talking heads.

cs.CV↗