Searcharxiv⌕ Search

arXiv subjects

Hao Gao

Publications and source records attributed to Hao Gao.

At least 19 recordsLinked to original sources
arXiv2026

Emulation strategies for Bayesian inference of regional left ventricle material parameters

Patient-specific biomechanical models of the left ventricle can relate cardiac magnetic resonance imaging to regional myocardial material properties, but existing emulator-based studies typically treat the myocardium as mechanically homogeneous, limiting representation of localised dysfunction. We propose a Bayesian surrogate-modelling framework for inferring regional Holzapfel-Ogden material parameters in a left ventricle partitioned into five physiological zones derived from the American Heart Association 17-segment model. Eight emulator strategies spanning single- versus multi-output, local versus global, and Gaussian-process- versus neural-network-based architectures were screened using parameter point-estimation accuracy; the three retained models were evaluated using empirical marginal credible-interval coverage and posterior contraction. We found that models with comparable point accuracy nevertheless differed markedly in uncertainty. A multi-output variational Gaussian process provided the most favourable balance across these criteria and was retained for the subsequent analyses. In synthetic local and global stiffening scenarios, maximum a posteriori estimates generally distinguished stiffened from baseline zones, but the nonlinear-stiffening parameters were more difficult to identify from end-diastolic observations than the stiffness-magnitude parameters. A healthy-volunteer analysis demonstrates feasibility with an incomplete observation vector and jointly inferred strain-noise scales. These results suggest that the proposed framework provides a computationally feasible, uncertainty-aware approach to regional left ventricle parameter inference.

stat.AP↗
arXiv2026

ETHead: Generating Expressive 3D Facial Animation and Head Movement from Speech

Generating expressive 3D talking heads solely from speech remains a significant challenge due to the scarcity of high-fidelity 3D data, which limits the modeling of complex emotional motion patterns. In this paper, we introduce \textbf{E}xpressive \textbf{T}alking \textbf{Head} (ETHead), a method for generating 3D facial and head motions that vividly align with the emotional content of input speech. To overcome the data limitations, we design a self-distillation framework that leverages large-scale 2D talking videos to pre-train a specialized speech encoder. By incorporating a novel emotion-modulated probabilistic masking mechanism, this framework aligns speech representations with expressive visual dynamics, allowing the encoder to extract features highly correlated with facial and head motions directly from audio. These features are then leveraged to guide 3D generation, enriching input cues and providing explicit supervision through a joint speech-motion latent space. Extensive experiments demonstrate that ETHead substantially outperforms state-of-the-art methods. Furthermore, our motion-aligned speech encoder can serve as a transferable module, offering a general solution for enhancing expressiveness in other 3D talking head animation frameworks. The project page is available at https://verdure-oss.github.io/ETHead.github.io/.

cs.GR↗
arXiv2026

NGM-RAG: Neural Graph Matching based Retrieval-Augmented Generation

Retrieval-Augmented Generation (RAG) significantly enhances the ability of Large Language Models (LLMs) to provide accurate and contextually relevant answers by dynamically integrating external databases. However, traditional RAG methods are primarily constrained by their reliance on text-based retrieval strategies, which often struggle with complex questions requiring multi-hop reasoning. To address this limitation, we introduce Neural Graph Matching based Retrieval-Augmented Generation (NGM-RAG), a novel framework that leverages graph structures to effectively capture and utilize relational knowledge for improved retrieval and answer generation. NGM-RAG explicitly incorporates graph construction, graph matching, and answer generation into a unified process. Within this framework, we propose a neural graph matching approach that combines text-based matching with Graph Neural Networks (GNNs). By employing an adaptive weighting strategy, NGM-RAG efficiently integrates multiple matching methods to select the most relevant contextual node information for answer generation. Experimental results on multi-hop question answering and long-context summarization tasks demonstrate that our NGM-RAG model achieves superior performance compared to both traditional NaiveRAG methods and state-of-the-art graph-enhanced approaches such as GraphRAG and LightRAG.

cs.IR↗
arXiv2026

Effective Gaussian Management for High-fidelity Scene Reconstruction

This paper proposes an effective Gaussian management framework for high-fidelity scene reconstruction of both appearance and geometry. Unlike recent Gaussian Splatting (GS) pipelines that treat all primitives uniformly during optimization, our framework explicitly manages the attribute activation, representation and pruning of Gaussian. Specifically, our framework first introduces GauSep, a novel densification strategy that selectively activates Gaussian color or normal attributes to alleviate destructive gradient conflicts arising from dual supervision. We further propose GauRep, an adaptive Gaussian representation that dynamically adjusts spherical harmonics (SHs) orders and performs task-decoupled pruning to reduce redundancy at both the individual and global levels. To provide reliable geometric supervision for above mangement process, we additionally introduce CoRe, an regularized surface reconstruction module that distills robust normal fields from an SDF branch to the Gaussian representation through a confidence mechanism. Notably, the proposed Gaussian management is compatible with various reconstruction architectures and can be seamlessly integrated to improve performance while reducing size of the model. Extensive experiments demonstrate that our approach achieves superior or comparable performance in appearance and geometry reconstruction compared with state-of-the-art methods, while using significantly fewer parameters.

cs.CV↗
arXiv2026

High-order tensor neural network for iteration-free structure relaxation

Structure relaxation is important for the discovery of new materials, yet conventional ab initio optimization remains a major bottleneck in high-throughput screening workflows. Machine learning potentials have accelerated relaxation by orders of magnitude, but they still rely on iterative optimization and high-quality DFT force labels. Here, we present HotRelax, a high-order tensor message-passing neural network for one-shot, end-to-end prediction of relaxed structures. Trained directly on paired unrelaxed and relaxed structures, HotRelax requires no DFT force labels and predicts relaxed structures in a single forward pass, without iterative inference or post-processing. Across five diverse datasets spanning 3D bulk crystals, 2D layered materials and catalysts, HotRelax shows strong performance relative to state-of-the-art end-to-end relaxation models, achieving lower prediction errors on several benchmarks while maintaining a compact model size and efficient inference. Extensive DFT calculations further show that the predicted structures are close in energy to their DFT-relaxed counterparts. When integrated into catalytic workflows, HotRelax also improves the accuracy and generalization of relaxed-state energy prediction models. Together, these results support HotRelax as an efficient and widely applicable framework for end-to-end structure relaxation, with strong potential to accelerate high-throughput materials discovery.

physics.comp-ph↗
arXiv2026

SEEM: Exploiting Black-Box Text Attacks to Manipulate Tool Selection

Tool learning has emerged as a powerful auxiliary mechanism that extends the capabilities of large language models (LLMs), enabling them to address complex tasks that demand real-time relevance or high-precision operations. However, beneath this strength lie significant security risks. Prior studies have primarily concentrated on corrupting the outputs of invoked tools, while largely overlooking the vulnerability of the tool selection process itself. To bridge this gap, we introduce a black-box, text-based attack that substantially increases the likelihood of a target tool being selected. We propose SEEM, a two-level coarse-to-fine perturbation method that operates at both the word and character levels. Through comprehensive experiments, we show that merely perturbing the textual information of tools can markedly raise the probability of the target tool being prioritized and ranked higher among candidates. Our findings expose critical weaknesses in the tool selection mechanism and lay the groundwork for developing defenses to secure this essential process.

cs.CR↗
arXiv2026

Img2CADSeq: Image-to-CAD Generation via Sequence-Based Diffusion

Boundary Representation (BRep) is the standard format for Computer-Aided Design (CAD), yet reconstructing high-quality BReps from single-view images remains challenging due to the complexity of topological constraints and operation sequences. We present Img2CADSeq, a multi-stage pipeline that overcomes these limitations by encoding CAD sequences into a three-level hierarchical codebook. Guided by an importance prioritization, this strategy values profiles over details, compressing long sequences into a stable discrete latent space. To bridge the modality gap, we leverage a coarse-to-fine point cloud intermediate, aligning 2D visual features with 3D CAD sequences via contrastive learning to condition a VQ-Diffusion model. Supported by newly introduced CAD-220K and PrintCAD datasets, our approach ensures robust industrial domain adaptation. Extensive experiments demonstrate that Img2CADSeq significantly outperforms state-of-the-art methods, producing standard STEP files that can be directly used in commercial CAD software.

cs.CV↗
arXiv2026

Instance-level Visual Active Tracking with Occlusion-Aware Planning

Visual Active Tracking (VAT) aims to control cameras to follow a target in 3D space, which is critical for applications like drone navigation and security surveillance. However, it faces two key bottlenecks in real-world deployment: confusion from visually similar distractors caused by insufficient instance-level discrimination and severe failure under occlusions due to the absence of active planning. To address these, we propose OA-VAT, a unified pipeline with three complementary modules. First, a training-free Instance-Aware Offline Prototype Initialization aggregates multi-view augmented features via DINOv3 to construct discriminative instance prototypes, mitigating distractor confusion. Second, an Online Prototype Enhancement Tracker enhances prototypes online and integrates a confidence-aware Kalman filter for stable tracking under appearance and motion changes. Third, an Occlusion-Aware Trajectory Planner, trained on our new Planning-20k dataset, uses conditional diffusion to generate obstacle-avoiding paths for occlusion recovery. Experiments demonstrate OA-VAT achieves 0.93 average SR on UnrealCV (+2.2% vs. SOTA TrackVLA), 90.8% average CAR on real-world datasets (+12.1% vs. SOTA GC-VAT), and 81.6% TSR on a DJI Tello drone. Running at 35 FPS on an RTX 3090, it delivers robust, real-time performance for practical deployment.

cs.CV↗
arXiv2026

VADv2: End-to-End Vectorized Autonomous Driving via Probabilistic Planning

Learning a human-like driving policy from large-scale driving demonstrations is promising, but the uncertainty and non-deterministic nature of planning make it challenging. Existing learning-based planning methods follow a deterministic paradigm to directly regress the action, failing to cope with the uncertainty problem. In this work, we propose a probabilistic planning model for end-to-end autonomous driving, termed VADv2. We resort to a probabilistic field function to model the mapping from the action space to the probabilistic distribution. Since the planning action space is a high-dimensional continuous spatiotemporal space and hard to tackle, we first discretize the planning action space to a large planning vocabulary and then tokenize the planning vocabulary into planning tokens. Planning tokens interact with scene tokens and output the probabilistic distribution of action. Mass driving demonstrations are leveraged to supervise the distribution. VADv2 achieves state-of-the-art closed-loop performance on the CARLA Town05 benchmark, significantly outperforming existing methods, and also leads the recent Bench2Drive benchmark. We further provide comprehensive evaluations on NAVSIM and a large-scale 3DGS-based benchmark, demonstrating its effectiveness in real-world applications. Code is available at https://github.com/hustvl/VAD.

cs.CV↗
arXiv2026

RAD-2: Scaling Reinforcement Learning in a Generator-Discriminator Framework

High-level autonomous driving requires motion planners capable of modeling multimodal future uncertainties while remaining robust in closed-loop interactions. Although diffusion-based planners are effective at modeling complex trajectory distributions, they often suffer from stochastic instabilities and the lack of corrective negative feedback when trained purely with imitation learning. To address these issues, we propose RAD-2, a unified generator-discriminator framework for closed-loop planning. Specifically, a diffusion-based generator is used to produce diverse trajectory candidates, while an RL-optimized discriminator reranks these candidates according to their long-term driving quality. This decoupled design avoids directly applying sparse scalar rewards to the full high-dimensional trajectory space, thereby improving optimization stability. To further enhance reinforcement learning, we introduce Temporally Consistent Group Relative Policy Optimization, which exploits temporal coherence to alleviate the credit assignment problem. In addition, we propose On-policy Generator Optimization, which converts closed-loop feedback into structured longitudinal optimization signals and progressively shifts the generator toward high-reward trajectory manifolds. To support efficient large-scale training, we introduce BEV-Warp, a high-throughput simulation environment that performs closed-loop evaluation directly in Bird's-Eye View feature space via spatial warping. RAD-2 reduces the collision rate by 56% compared with strong diffusion-based planners. Real-world deployment further demonstrates improved perceived safety and driving smoothness in complex urban traffic.

cs.CV↗
arXiv2026

Green AI Carbon Optimizer: Carbon-Efficient Training Location Recommendation and Global AI Energy Demand Forecasting

AI training and deployment consume substantial electricity, but carbon outcomes remain weakly integrated into routine model development decisions. This paper presents Green AI Carbon Optimizer with two primary contributions: (i) a carbon aware cloud region recommendation method for training workloads, and (ii) a power law forecasting pipeline for global AI energy demand. For location recommendation, we combine regional grid carbon intensity, renewable share, and data center Power Usage Effectiveness (PUE) into a unified scoring model across 100+ regions from major cloud providers. For a reference workload (8*A100, 100h), estimated emissions in our sampled regions range from 7.74kg to 272.00kg CO2. Selecting the best region instead of the worst corresponds to a 97.2% reduction relative to the worst case. Ablation shows that ranking by renewable share alone can select regions with higher CO2 emissions than rankings that include grid carbon intensity. For forecasting, we fit a power law relation between parameter count and training energy using 26 anchor models. We combine this fit with scenario assumptions on model growth, hardware efficiency, and training frequency, and evaluate sensitivity to inference ratio and ecosystem scaling. Across scenarios, projected 2030 demand ranges from 7TWh to 1,436TWh under the stated assumptions, highlighting the importance of deployment choices, model scaling discipline, and transparent energy reporting.

cs.PF↗
arXiv2026

Anharmonicity Driven by Vacancy Ordering Unlocks High-performance Thermoelectric Conversion in Defective Chalcopyrites II-III$_2$-VI$_4$

Defective chalcopyrites have recently emerged as promising thermoelectric materials because their ordered intrinsic vacancies can profoundly reshape both lattice dynamics and electronic structure. Here, we present a comprehensive theoretical investigation of the lattice thermal and carrier transport properties of II-III$_2$-VI$_4$ defective chalcopyrites by combining first-principles calculations with machine-learning interatomic potentials. We show that vacancy ordering enhances lattice distortion, leading to strong anharmonicity and metavalent bonding. The interplay of soft low-frequency phonons, strongly negative Grüneisen parameters, and a substantially enlarged four-phonon scattering phase space results in four-phonon-scattering-dominated heat transport, yielding ultralow lattice thermal conductivity. Meanwhile, systematic anion substitution at the VI-site provides an effective route to tune the electronic structure: reduced anion electronegativity weakens metal-anion hybridization, shifts anion $p$ states upward, narrows the band gap, and thereby improves electrical transport. Benefiting from this synergy between vacancy-induced phonon suppression and anion-regulated electronic optimization, CdGa$_2$Te$_4$ exhibits an ultralow lattice thermal conductivity of 0.19 W$\cdot$m$^{-1}$K$^{-1}$ and a high room-temperature $ZT$ of 0.957. This work not only predicts defective chalcopyrites as a promising platform for high-performance thermoelectrics but also provides a practical design strategy by integrating vacancy ordering, higher-order phonon scattering, and anion-dependent band engineering.

cond-mat.mtrl-sci↗
arXiv2026

Senna-2: Aligning VLM and End-to-End Driving Policy for Consistent Decision Making and Planning

Vision-language models (VLMs) enhance the planning capability of end-to-end (E2E) driving policy by leveraging high-level semantic reasoning. However, existing approaches often overlook the dual-system consistency between VLM's high-level decision and E2E's low-level planning. As a result, the generated trajectories may misalign with the intended driving decisions, leading to weakened top-down guidance and decision-following ability of the system. To address this issue, we propose Senna-2, an advanced VLM-E2E driving policy that explicitly aligns the two systems for consistent decision-making and planning. Our method follows a consistency-oriented three-stage training paradigm. In the first stage, we conduct driving pre-training to achieve preliminary decision-making and planning, with a decision adapter transmitting VLM decisions to E2E policy in the form of implicit embeddings. In the second stage, we align the VLM and the E2E policy in an open-loop setting. In the third stage, we perform closed-loop alignment via bottom-up Hierarchical Reinforcement Learning in 3DGS environments to reinforce the safety and efficiency. Extensive experiments demonstrate that Senna-2 achieves superior dual-system consistency (19.3% F1 score improvement) and significantly enhances driving safety in both open-loop (5.7% FDE reduction) and closed-loop settings (30.6% AF-CR reduction).

cs.CV↗
arXiv2026

Low-SWaP Magneto-optical Trap using both Planar Optical and Magnetic Components

Compact, lightweight, and energy-efficient cold atom systems are crucial for advancing quantum technologies, yet their realization remains constrained by the bulky optical and magnetic components required in current atom trapping architectures. Here, we demonstrate a low-SWaP magneto-optical trap that seamlessly integrates planar optical and magnetic components into a unified platform. A monolithic dual-functional metasurface simultaneously polarized and shapes the cooling beam, replacing traditional lens-waveplate assemblies and converting a linearly polarized Gaussian beam into a circularly polarized flat-top beam. In parallel, a planar coil chip substitutes bulky anti-Helmholtz cols and generated the required quadrupole magnetic field with drastically reduced power consumption. Under D2 line cooling of 87Rb atoms, the fully planar system delivers nearly an order-of-magnitude improvement in trapping performance while operating at a fraction of the size, weight, and power of traditional systems. This compact, bulky-component-free approach offers a scalable, energy-efficient pathway toward chip-scale cold atom platforms.

physics.atom-ph↗
arXiv2025

Iterative learning scheme for crystal structure prediction with anharmonic lattice dynamics

First-principles based crystal structure prediction (CSP) methods have revealed an essential tool for the discovery of new materials. However, in solids close to displacive phase transitions, which are common in ferroelectrics, thermoelectrics, charge-density wave systems, or superconducting hydrides, the ionic contribution to the free energy and lattice anharmonicity become essential, limiting the capacity of CSP techniques to determine the thermodynamical stability of competing phases. While variational methods like the stochastic self-consistent harmonic approximation (SSCHA) accurately account for anharmonic lattice dynamics \emph{ab initio}, their high computational cost makes them impractical for CSP. Machine-learning interatomic potentials offer accelerated sampling of the energy landscape compared to purely first-principles approaches, but their reliance on extensive training data and limited generalization restricts practical applications. Here, we propose an iterative learning framework combining evolutionary algorithms, atomic foundation models, and SSCHA to enable CSP with anharmonic lattice dynamics. Foundation models enable robust relaxations of random structures, drastically reducing required training data. Applied to the highly anharmonic H$_3$S system, our framework achieves good agreement with the benchmarks based on density functional theory, accurately predicting phase stability and vibrational properties from 50 to 200 GPa. Importantly, we find that the statistical averaging in the SSCHA reduces the error in the free energy evaluation, avoiding the need for extremely high accuracy of machine-learning potentials. This approach bridges the gap between data efficiency and predictive power, establishing a practical pathway for CSP with anharmonic lattice dynamics.

cond-mat.mtrl-sci↗
arXiv2025

RAD: Training an End-to-End Driving Policy via Large-Scale 3DGS-based Reinforcement Learning

Existing end-to-end autonomous driving (AD) algorithms typically follow the Imitation Learning (IL) paradigm, which faces challenges such as causal confusion and an open-loop gap. In this work, we propose RAD, a 3DGS-based closed-loop Reinforcement Learning (RL) framework for end-to-end Autonomous Driving. By leveraging 3DGS techniques, we construct a photorealistic digital replica of the real physical world, enabling the AD policy to extensively explore the state space and learn to handle out-of-distribution scenarios through large-scale trial and error. To enhance safety, we design specialized rewards to guide the policy in effectively responding to safety-critical events and understanding real-world causal relationships. To better align with human driving behavior, we incorporate IL into RL training as a regularization term. We introduce a closed-loop evaluation benchmark consisting of diverse, previously unseen 3DGS environments. Compared to IL-based methods, RAD achieves stronger performance in most closed-loop metrics, particularly exhibiting a 3x lower collision rate. Abundant closed-loop results are presented in the supplementary material. Code is available at https://github.com/hustvl/RAD for facilitating future research.

cs.CV↗
arXiv2025

Minibeam-pLATTICE: A novel proton LATTICE modality using minibeams

Purpose: LATTICE, a form of spatially fractionated radiation therapy that delivers high-dose peaks and low-dose valleys within the target, has been clinically utilized for treating bulky tumors. However, its application to small-to-medium-sized target remains challenging due to beam size limitations. To address this challenge, this work proposes a novel proton LATTICE (pLATTICE) modality using minibeams, namely minibeam-pLATTICE, that extends LATTICE approach for small-to-medium targets. Methods: Three minibeam-pLATTICE methods are introduced. (1) M0: a fixed minibeam orientation for all beam angles; (2) M1: alternated minibeam orientations, for consecutive beam angles; (3) M2: multiple minibeam orientations for each beam angle. For each minibeam-pLATTICE method, an optimization problem is formulated to optimize dose uniformity in target peaks and valleys, as well as dose-volume-histogram-based objectives. This problem is solved using iterative convex relaxation and alternating direction method of multipliers. Results: Three minibeam-pLATTICE methods are validated to demonstrate the feasibility of minibeam-pLATTICE for head-and-neck cases. The advantages of this modality over conventional beam (CONV) pLATTICE are evaluated by comparing peak-to-valley dose ratio (PVDR) and dose delivered to organs at risk (OAR). All three minibeam-pLATTICE modalities achieved improved plan quality compared to CONV, with M2 yielding the best results. For example, in terms of PVDR, M2=5.89, compared to CONV=4.13, M0=4.87 and M1=4.7. Conclusion: A novel minibeam-pLATTICE modality is proposed that generates lattice dose patterns for small-to-medium targets, which are not achievable with conventional pLATTICE due to beam size limitations.

physics.med-ph↗
arXiv2025

Optimizing normal tissue sparing via spatiotemporal optimization under equivalent tumor-radical efficacy

Objective: Spatiotemporal optimization in radiation therapy involves determining the optimal number of dose delivery fractions (temporal) and the optimal dose per fraction (spatial). Traditional approaches focus on maximizing the biologically effective dose (BED) to the target while constraining BED to organs-at-risk (OAR), which may lead to insufficient BED for complete tumor cell kill. This work proposes a formulation that ensures adequate BED delivery to the target while minimizing BED to the OAR. Approach: A spatiotemporal optimization model is developed that incorporates an inequality constraint to guarantee sufficient BED for tumor cell kill while minimizing BED to the OAR. The model accounts for tumor proliferation dynamics, including lag time (delay before proliferation begins) and doubling time (time for tumor volume to double), to optimize dose fractionation. Results: The performance of our formulation is evaluated for varying lag and doubling times. The results show that mean BED to the target consistently meets the minimum requirement for tumor cell kill. Additionally, the mean BED to OAR varies based on tumor proliferation dynamics. In the prostate case with lag time of 7 days and doubling time of 2 days, it is observed that mean BED delivered to femoral head is lowest at around 20 fractions, making this an optimal choice. While in the head-and-neck case, mean BED to OAR decreases as the number of fractions increases, suggesting that a higher number of fractions is optimal. Significance: A spatiotemporal optimization model is presented that minimizes BED to the OAR while ensuring sufficient BED for tumor cell kill. By incorporating tumor lag and doubling time, the approach identifies optimal number of fractions. This model can be extended to support hyperfractionation or accelerated fractionation strategies, offering a versatile tool for clinical treatment planning.

physics.med-ph↗