arXiv subjects

He Wang

Publications and source records attributed to He Wang.

At least 163 records · Page 9Linked to original sources

arXiv2025

OctoNav: Towards Generalist Embodied Navigation

Embodied navigation stands as a foundation pillar within the broader pursuit of embodied AI. However, previous navigation research is divided into different tasks/capabilities, e.g., ObjNav, ImgNav and VLN, where they differ in task objectives and modalities, making datasets and methods are designed individually. In this work, we take steps toward generalist navigation agents, which can follow free-form instructions that include arbitrary compounds of multi-modal and multi-capability. To achieve this, we propose a large-scale benchmark and corresponding method, termed OctoNav-Bench and OctoNav-R1. Specifically, OctoNav-Bench features continuous environments and is constructed via a designed annotation pipeline. We thoroughly craft instruction-trajectory pairs, where instructions are diverse in free-form with arbitrary modality and capability. Also, we construct a Think-Before-Action (TBA-CoT) dataset within OctoNav-Bench to provide the thinking process behind actions. For OctoNav-R1, we build it upon MLLMs and adapt it to a VLA-type model, which can produce low-level actions solely based on 2D visual observations. Moreover, we design a Hybrid Training Paradigm (HTP) that consists of three stages, i.e., Action-/TBA-SFT, Nav-GPRO, and Online RL stages. Each stage contains specifically designed learning policies and rewards. Importantly, for TBA-SFT and Nav-GRPO designs, we are inspired by the OpenAI-o1 and DeepSeek-R1, which show impressive reasoning ability via thinking-before-answer. Thus, we aim to investigate how to achieve thinking-before-action in the embodied navigation field, to improve model's reasoning ability toward generalists. Specifically, we propose TBA-SFT to utilize the TBA-CoT dataset to fine-tune the model as a cold-start phrase and then leverage Nav-GPRO to improve its thinking ability. Finally, OctoNav-R1 shows superior performance compared with previous methods.

cs.CV↗

arXiv2025

The future of gravitational wave science unlocking LIGO potential: AI-driven data analysis and exploration

The advent of gravitational wave astronomy (GW) has revolutionized the observation of cataclysmic cosmic events, such as black hole mergers and neutron star collisions. The Laser Interferometer Gravitational-Wave Observatory (LIGO) has been at the forefront of these discoveries. However, the immense volume and complexity of gravitational wave data present significant challenges for traditional analysis methods. This paper investigates the growing synergy between artificial intelligence (AI) and GW science, emphasizing how AI enhances signal detection, noise reduction, and data interpretation. It begins with an overview of GW fundamentals and the role of machine learning in increasing detector sensitivity. Notable GW events observed by LIGO are discussed alongside persistent analytical challenges such as data quality, generalization, and computational constraints. A comprehensive performance review of AI techniques, including supervised learning, unsupervised learning, deep learning, and reinforcement learning, is presented based on data spanning 2021 to 2024. Evaluation metrics include accuracy, precision, true positive rate (TPR), false positive rate (FPR), and computational efficiency. Findings indicate that deep learning and supervised learning outperform other approaches, particularly in enhancing TPR and minimizing FPR. While unsupervised and reinforcement learning models offer less precision, they demonstrate high efficiency and potential for real-time applications. The study also explores AI integration into next-generation detectors and waveform reconstruction techniques. Overall, the integration of AI into GW research significantly improves the reliability and speed of event detection, unlocking new possibilities for exploring the dynamic universe. This paper provides a comprehensive outlook on the transformative role of AI in shaping the future of GW astronomy.

astro-ph.GA↗

arXiv2025

Accelerating Bayesian Sampling for Massive Black Hole Binaries with Prior Constraints from Conditional Variational Autoencoder

A Conditional Variational Autoencoder (CVAE) model is employed for parameter inference on gravitational waves (GW) signals of massive black hole binaries, considering joint observations with a network of three space-based GW detectors. Our experiments show that the trained CVAE model can estimate the posterior distribution of source parameters in approximately one second, while the standard Bayesian sampling method, utilizing parallel computation across 16 CPU cores, takes an average of 20 hours for a GW signal instance. However, the sampling distributions from CVAE exhibit lighter tails, appearing broader when compared to the standard Bayesian sampling results. By using CVAE results to constrain the prior range for Bayesian sampling, the sampling time is reduced by a factor of $\sim$6 while maintaining the similar precision of the Bayesian results.

astro-ph.IM↗

arXiv2025

Adversarially Robust AI-Generated Image Detection for Free: An Information Theoretic Perspective

Rapid advances in Artificial Intelligence Generated Images (AIGI) have facilitated malicious use, such as forgery and misinformation. Therefore, numerous methods have been proposed to detect fake images. Although such detectors have been proven to be universally vulnerable to adversarial attacks, defenses in this field are scarce. In this paper, we first identify that adversarial training (AT), widely regarded as the most effective defense, suffers from performance collapse in AIGI detection. Through an information-theoretic lens, we further attribute the cause of collapse to feature entanglement, which disrupts the preservation of feature-label mutual information. Instead, standard detectors show clear feature separation. Motivated by this difference, we propose Training-free Robust Detection via Information-theoretic Measures (TRIM), the first training-free adversarial defense for AIGI detection. TRIM builds on standard detectors and quantifies feature shifts using prediction entropy and KL divergence. Extensive experiments across multiple datasets and attacks validate the superiority of our TRIM, e.g., outperforming the state-of-the-art defense by 33.88% (28.91%) on ProGAN (GenImage), while well maintaining original accuracy.

cs.CV↗

arXiv2025

AISHELL-5: The First Open-Source In-Car Multi-Channel Multi-Speaker Speech Dataset for Automatic Speech Diarization and Recognition

This paper delineates AISHELL-5, the first open-source in-car multi-channel multi-speaker Mandarin automatic speech recognition (ASR) dataset. AISHLL-5 includes two parts: (1) over 100 hours of multi-channel speech data recorded in an electric vehicle across more than 60 real driving scenarios. This audio data consists of four far-field speech signals captured by microphones located on each car door, as well as near-field signals obtained from high-fidelity headset microphones worn by each speaker. (2) a collection of 40 hours of real-world environmental noise recordings, which supports the in-car speech data simulation. Moreover, we also provide an open-access, reproducible baseline system based on this dataset. This system features a speech frontend model that employs speech source separation to extract each speaker's clean speech from the far-field signals, along with a speech recognition module that accurately transcribes the content of each individual speaker. Experimental results demonstrate the challenges faced by various mainstream ASR models when evaluated on the AISHELL-5. We firmly believe the AISHELL-5 dataset will significantly advance the research on ASR systems under complex driving scenarios by establishing the first publicly available in-car ASR benchmark.

cs.SD↗

arXiv2025

TrackVLA: Embodied Visual Tracking in the Wild

Embodied visual tracking is a fundamental skill in Embodied AI, enabling an agent to follow a specific target in dynamic environments using only egocentric vision. This task is inherently challenging as it requires both accurate target recognition and effective trajectory planning under conditions of severe occlusion and high scene dynamics. Existing approaches typically address this challenge through a modular separation of recognition and planning. In this work, we propose TrackVLA, a Vision-Language-Action (VLA) model that learns the synergy between object recognition and trajectory planning. Leveraging a shared LLM backbone, we employ a language modeling head for recognition and an anchor-based diffusion model for trajectory planning. To train TrackVLA, we construct an Embodied Visual Tracking Benchmark (EVT-Bench) and collect diverse difficulty levels of recognition samples, resulting in a dataset of 1.7 million samples. Through extensive experiments in both synthetic and real-world environments, TrackVLA demonstrates SOTA performance and strong generalizability. It significantly outperforms existing methods on public benchmarks in a zero-shot manner while remaining robust to high dynamics and occlusion in real-world scenarios at 10 FPS inference speed. Our project page is: https://pku-epic.github.io/TrackVLA-web.

cs.RO↗

arXiv2025

ASGrasp: Generalizable Transparent Object Reconstruction and 6-DoF Grasp Detection from RGB-D Active Stereo Camera

In this paper, we tackle the problem of grasping transparent and specular objects. This issue holds importance, yet it remains unsolved within the field of robotics due to failure of recover their accurate geometry by depth cameras. For the first time, we propose ASGrasp, a 6-DoF grasp detection network that uses an RGB-D active stereo camera. ASGrasp utilizes a two-layer learning-based stereo network for the purpose of transparent object reconstruction, enabling material-agnostic object grasping in cluttered environments. In contrast to existing RGB-D based grasp detection methods, which heavily depend on depth restoration networks and the quality of depth maps generated by depth cameras, our system distinguishes itself by its ability to directly utilize raw IR and RGB images for transparent object geometry reconstruction. We create an extensive synthetic dataset through domain randomization, which is based on GraspNet-1Billion. Our experiments demonstrate that ASGrasp can achieve over 90% success rate for generalizable transparent object grasping in both simulation and the real via seamless sim-to-real transfer. Our method significantly outperforms SOTA networks and even surpasses the performance upper bound set by perfect visible point cloud inputs.Project page: https://pku-epic.github.io/ASGrasp

cs.RO↗

arXiv2025

Recent Deep Learning in Crowd Behaviour Analysis: A Brief Review

Crowd behaviour analysis is essential to numerous real-world applications, such as public safety and urban planning, and therefore has been studied for decades. In the last decade or so, the development of deep learning has significantly propelled the research on crowd behaviours. This chapter reviews recent advances in crowd behaviour analysis using deep learning. We mainly review the research in two core tasks in this field, crowd behaviour prediction and recognition. We broadly cover how different deep neural networks, after first being proposed in machine learning, are applied to analysing crowd behaviours. This includes pure deep neural network models as well as recent development of methodologies combining physics with deep learning. In addition, representative studies are discussed and compared in detail. Finally, we discuss the effectiveness of existing methods and future research directions in this rapidly evolving field. This chapter aims to provide a high-level summary of the ongoing deep learning research in crowd behaviour analysis. It intends to help new researchers who just entered this field to obtain an overall understanding of the ongoing research, as well as to provide a retrospective analysis for existing researchers to identify possible future directions

cs.CV↗

arXiv2025

Large-Scale Multi-Character Interaction Synthesis

Generating large-scale multi-character interactions is a challenging and important task in character animation. Multi-character interactions involve not only natural interactive motions but also characters coordinated with each other for transition. For example, a dance scenario involves characters dancing with partners and also characters coordinated to new partners based on spatial and temporal observations. We term such transitions as coordinated interactions and decompose them into interaction synthesis and transition planning. Previous methods of single-character animation do not consider interactions that are critical for multiple characters. Deep-learning-based interaction synthesis usually focuses on two characters and does not consider transition planning. Optimization-based interaction synthesis relies on manually designing objective functions that may not generalize well. While crowd simulation involves more characters, their interactions are sparse and passive. We identify two challenges to multi-character interaction synthesis, including the lack of data and the planning of transitions among close and dense interactions. Existing datasets either do not have multiple characters or do not have close and dense interactions. The planning of transitions for multi-character close and dense interactions needs both spatial and temporal considerations. We propose a conditional generative pipeline comprising a coordinatable multi-character interaction space for interaction synthesis and a transition planning network for coordinations. Our experiments demonstrate the effectiveness of our proposed pipeline for multicharacter interaction synthesis and the applications facilitated by our method show the scalability and transferability.

cs.GR↗

arXiv2025

DSMentor: Enhancing Data Science Agents with Curriculum Learning and Online Knowledge Accumulation

Large language model (LLM) agents have shown promising performance in generating code for solving complex data science problems. Recent studies primarily focus on enhancing in-context learning through improved search, sampling, and planning techniques, while overlooking the importance of the order in which problems are tackled during inference. In this work, we develop a novel inference-time optimization framework, referred to as DSMentor, which leverages curriculum learning -- a strategy that introduces simpler task first and progressively moves to more complex ones as the learner improves -- to enhance LLM agent performance in challenging data science tasks. Our mentor-guided framework organizes data science tasks in order of increasing difficulty and incorporates a growing long-term memory to retain prior experiences, guiding the agent's learning progression and enabling more effective utilization of accumulated knowledge. We evaluate DSMentor through extensive experiments on DSEval and QRData benchmarks. Experiments show that DSMentor using Claude-3.5-Sonnet improves the pass rate by up to 5.2% on DSEval and QRData compared to baseline agents. Furthermore, DSMentor demonstrates stronger causal reasoning ability, improving the pass rate by 8.8% on the causality problems compared to GPT-4 using Program-of-Thoughts prompts. Our work underscores the importance of developing effective strategies for accumulating and utilizing knowledge during inference, mirroring the human learning process and opening new avenues for improving LLM performance through curriculum-based inference optimization.

cs.AI↗

arXiv2025

Deep learning-driven likelihood-free parameter inference for 21-cm forest observations

The hyperfine structure absorption lines of neutral hydrogen in spectra of high-redshift radio sources, known collectively as the 21-cm forest, have been demonstrated as a sensitive probe to the small-scale structures governed by the dark matter (DM) properties, as well as the thermal history of the intergalactic medium regulated by the first galaxies during the epoch of reionization. By statistically analyzing these spectral features, the one-dimensional (1D) power spectrum of the 21-cm forest can effectively break the parameter degeneracies and constrain the properties of both DM and the first galaxies. However, conventional parameter inference methods face challenges due to computationally expensive simulations for 21-cm forest and the non-Gaussian signal characteristics. To address these issues, we introduce generative normalizing flows for data augmentation and inference normalizing flows for parameters estimation. This approach efficiently estimates parameters from minimally simulated datasets with non-Gaussian signals. Using simulated data from the upcoming Square Kilometre Array (SKA), we demonstrate the ability of the deep learning-driven likelihood-free approach to generate accurate posterior distributions, providing a robust and efficient tool for probing DM and the cosmic heating history using the 1D power spectrum of 21-cm forest in the era of SKA. This methodology is adaptable for scientific analyses with other unevenly distributed data.

astro-ph.CO↗

arXiv2025

Interactive Rendering of Relightable and Animatable Gaussian Avatars

Creating relightable and animatable avatars from multi-view or monocular videos is a challenging task for digital human creation and virtual reality applications. Previous methods rely on neural radiance fields or ray tracing, resulting in slow training and rendering processes. By utilizing Gaussian Splatting, we propose a simple and efficient method to decouple body materials and lighting from sparse-view or monocular avatar videos, so that the avatar can be rendered simultaneously under novel viewpoints, poses, and lightings at interactive frame rates (6.9 fps). Specifically, we first obtain the canonical body mesh using a signed distance function and assign attributes to each mesh vertex. The Gaussians in the canonical space then interpolate from nearby body mesh vertices to obtain the attributes. We subsequently deform the Gaussians to the posed space using forward skinning, and combine the learnable environment light with the Gaussian attributes for shading computation. To achieve fast shadow modeling, we rasterize the posed body mesh from dense viewpoints to obtain the visibility. Our approach is not only simple but also fast enough to allow interactive rendering of avatar animation under environmental light changes. Experiments demonstrate that, compared to previous works, our method can render higher quality results at a faster speed on both synthetic and real datasets.

cs.CV↗

arXiv2025

Self-Learning Hyperspectral and Multispectral Image Fusion via Adaptive Residual Guided Subspace Diffusion Model

Hyperspectral and multispectral image (HSI-MSI) fusion involves combining a low-resolution hyperspectral image (LR-HSI) with a high-resolution multispectral image (HR-MSI) to generate a high-resolution hyperspectral image (HR-HSI). Most deep learning-based methods for HSI-MSI fusion rely on large amounts of hyperspectral data for supervised training, which is often scarce in practical applications. In this paper, we propose a self-learning Adaptive Residual Guided Subspace Diffusion Model (ARGS-Diff), which only utilizes the observed images without any extra training data. Specifically, as the LR-HSI contains spectral information and the HR-MSI contains spatial information, we design two lightweight spectral and spatial diffusion models to separately learn the spectral and spatial distributions from them. Then, we use these two models to reconstruct HR-HSI from two low-dimensional components, i.e, the spectral basis and the reduced coefficient, during the reverse diffusion process. Furthermore, we introduce an Adaptive Residual Guided Module (ARGM), which refines the two components through a residual guided function at each sampling step, thereby stabilizing the sampling process. Extensive experimental results demonstrate that ARGS-Diff outperforms existing state-of-the-art methods in terms of both performance and computational efficiency in the field of HSI-MSI fusion. Code is available at https://github.com/Zhu1116/ARGS-Diff.

cs.CV↗

arXiv2025

FlexSpeech: Towards Stable, Controllable and Expressive Text-to-Speech

Current speech generation research can be categorized into two primary classes: non-autoregressive and autoregressive. The fundamental distinction between these approaches lies in the duration prediction strategy employed for predictable-length sequences. The NAR methods ensure stability in speech generation by explicitly and independently modeling the duration of each phonetic unit. Conversely, AR methods employ an autoregressive paradigm to predict the compressed speech token by implicitly modeling duration with Markov properties. Although this approach improves prosody, it does not provide the structural guarantees necessary for stability. To simultaneously address the issues of stability and naturalness in speech generation, we propose FlexSpeech, a stable, controllable, and expressive TTS model. The motivation behind FlexSpeech is to incorporate Markov dependencies and preference optimization directly on the duration predictor to boost its naturalness while maintaining explicit modeling of the phonetic units to ensure stability. Specifically, we decompose the speech generation task into two components: an AR duration predictor and a NAR acoustic model. The acoustic model is trained on a substantial amount of data to learn to render audio more stably, given reference audio prosody and phone durations. The duration predictor is optimized in a lightweight manner for different stylistic variations, thereby enabling rapid style transfer while maintaining a decoupled relationship with the specified speaker timbre. Experimental results demonstrate that our approach achieves SOTA stability and naturalness in zero-shot TTS. More importantly, when transferring to a specific stylistic domain, we can accomplish lightweight optimization of the duration module solely with about 100 data samples, without the need to adjust the acoustic model, thereby enabling rapid and stable style transfer.

eess.AS↗

arXiv2025

A Comparative Review of RNA Language Models

Given usefulness of protein language models (LMs) in structure and functional inference, RNA LMs have received increased attentions in the last few years. However, these RNA models are often not compared against the same standard. Here, we divided RNA LMs into three classes (pretrained on multiple RNA types (especially noncoding RNAs), specific-purpose RNAs, and LMs that unify RNA with DNA or proteins or both) and compared 13 RNA LMs along with 3 DNA and 1 protein LMs as controls in zero-shot prediction of RNA secondary structure and functional classification. Results shows that the models doing well on secondary structure prediction often perform worse in function classification or vice versa, suggesting that more balanced unsupervised training is needed.

q-bio.BM↗

arXiv2025

3D Student Splatting and Scooping

Recently, 3D Gaussian Splatting (3DGS) provides a new framework for novel view synthesis, and has spiked a new wave of research in neural rendering and related applications. As 3DGS is becoming a foundational component of many models, any improvement on 3DGS itself can bring huge benefits. To this end, we aim to improve the fundamental paradigm and formulation of 3DGS. We argue that as an unnormalized mixture model, it needs to be neither Gaussians nor splatting. We subsequently propose a new mixture model consisting of flexible Student's t distributions, with both positive (splatting) and negative (scooping) densities. We name our model Student Splatting and Scooping, or SSS. When providing better expressivity, SSS also poses new challenges in learning. Therefore, we also propose a new principled sampling approach for optimization. Through exhaustive evaluation and comparison, across multiple datasets, settings, and metrics, we demonstrate that SSS outperforms existing methods in terms of quality and parameter efficiency, e.g. achieving matching or better quality with similar numbers of components, and obtaining comparable results while reducing the component number by as much as 82%.

cs.CV↗

arXiv2025

High-fidelity 3D Object Generation from Single Image with RGBN-Volume Gaussian Reconstruction Model

Recently single-view 3D generation via Gaussian splatting has emerged and developed quickly. They learn 3D Gaussians from 2D RGB images generated from pre-trained multi-view diffusion (MVD) models, and have shown a promising avenue for 3D generation through a single image. Despite the current progress, these methods still suffer from the inconsistency jointly caused by the geometric ambiguity in the 2D images, and the lack of structure of 3D Gaussians, leading to distorted and blurry 3D object generation. In this paper, we propose to fix these issues by GS-RGBN, a new RGBN-volume Gaussian Reconstruction Model designed to generate high-fidelity 3D objects from single-view images. Our key insight is a structured 3D representation can simultaneously mitigate the afore-mentioned two issues. To this end, we propose a novel hybrid Voxel-Gaussian representation, where a 3D voxel representation contains explicit 3D geometric information, eliminating the geometric ambiguity from 2D images. It also structures Gaussians during learning so that the optimization tends to find better local optima. Our 3D voxel representation is obtained by a fusion module that aligns RGB features and surface normal features, both of which can be estimated from 2D images. Extensive experiments demonstrate the superiority of our methods over prior works in terms of high-quality reconstruction results, robust generalization, and good efficiency.

cs.CV↗

arXiv2025

Code-as-Monitor: Constraint-aware Visual Programming for Reactive and Proactive Robotic Failure Detection

Automatic detection and prevention of open-set failures are crucial in closed-loop robotic systems. Recent studies often struggle to simultaneously identify unexpected failures reactively after they occur and prevent foreseeable ones proactively. To this end, we propose Code-as-Monitor (CaM), a novel paradigm leveraging the vision-language model (VLM) for both open-set reactive and proactive failure detection. The core of our method is to formulate both tasks as a unified set of spatio-temporal constraint satisfaction problems and use VLM-generated code to evaluate them for real-time monitoring. To enhance the accuracy and efficiency of monitoring, we further introduce constraint elements that abstract constraint-related entities or their parts into compact geometric elements. This approach offers greater generality, simplifies tracking, and facilitates constraint-aware visual programming by leveraging these elements as visual prompts. Experiments show that CaM achieves a 28.7% higher success rate and reduces execution time by 31.8% under severe disturbances compared to baselines across three simulators and a real-world setting. Moreover, CaM can be integrated with open-loop control policies to form closed-loop systems, enabling long-horizon tasks in cluttered scenes with dynamic environments.

cs.RO↗