arXiv subjects

Hao Huang

Publications and source records attributed to Hao Huang.

At least 91 records · Page 5Linked to original sources

arXiv2025

Whistle: Data-Efficient Multilingual and Crosslingual Speech Recognition via Weakly Phonetic Supervision

There exist three approaches for multilingual and crosslingual automatic speech recognition (MCL-ASR) - supervised pretraining with phonetic or graphemic transcription, and self-supervised pretraining. We find that pretraining with phonetic supervision has been underappreciated so far for MCL-ASR, while conceptually it is more advantageous for information sharing between different languages. This paper explores the approach of pretraining with weakly phonetic supervision towards data-efficient MCL-ASR, which is called Whistle. We relax the requirement of gold-standard human-validated phonetic transcripts, and obtain International Phonetic Alphabet (IPA) based transcription by leveraging the LanguageNet grapheme-to-phoneme (G2P) models. We construct a common experimental setup based on the CommonVoice dataset, called CV-Lang10, with 10 seen languages and 2 unseen languages. A set of experiments are conducted on CV-Lang10 to compare, as fair as possible, the three approaches under the common setup for MCL-ASR. Experiments demonstrate the advantages of phoneme-based models (Whistle) for MCL-ASR, in terms of speech recognition for seen languages, crosslingual performance for unseen languages with different amounts of few-shot data, overcoming catastrophic forgetting, and training efficiency. It is found that when training data is more limited, phoneme supervision can achieve better results compared to subword supervision and self-supervision, thereby providing higher data-efficiency. To support reproducibility and promote future research along this direction, we release the code, models and data for the entire pipeline of Whistle at https://github.com/thu-spmi/CAT/tree/master/egs/cv-lang10.

cs.SD↗

arXiv2025

Socially-Aware Robot Navigation Enhanced by Bidirectional Natural Language Conversations Using Large Language Models

Robot navigation is crucial across various domains, yet traditional methods focus on efficiency and obstacle avoidance, often overlooking human behavior in shared spaces. With the rise of service robots, socially aware navigation has gained prominence. However, existing approaches primarily predict pedestrian movements or issue alerts, lacking true human-robot interaction. We introduce Hybrid Soft Actor-Critic with Large Language Model (HSAC-LLM), a novel framework for socially aware navigation. By integrating deep reinforcement learning with large language models, HSAC-LLM enables bidirectional natural language interactions, predicting both continuous and discrete navigation actions. When potential collisions arise, the robot proactively communicates with pedestrians to determine avoidance strategies. Experiments in 2D simulation, Gazebo, and real-world environments demonstrate that HSAC-LLM outperforms state-of-the-art DRL methods in interaction, navigation, and obstacle avoidance. This paradigm advances effective human-robot interactions in dynamic settings. Videos are available at https://hsacllm.github.io/.

cs.RO↗

arXiv2025

Computational identification of ketone metabolism as a key regulator of sleep stability and circadian dynamics via real-time metabolic profiling

Metabolism plays a crucial role in sleep regulation, yet its effects are challenging to track in real time. This study introduces a machine learning-based framework to analyze sleep patterns and identify how metabolic changes influence sleep at specific time points. We first established that sleep periods in Drosophila melanogaster function independently, with no causal relationship between different sleep episodes. Using gradient boosting models and explainable artificial intelligence techniques, we quantified the influence of time-dependent sleep features. Causal inference and autocorrelation analyses further confirmed that sleep states at different times are statistically independent, providing a robust foundation for exploring metabolic effects on sleep. Applying this framework to flies with altered monocarboxylate transporter 2 expression, we found that changes in ketone transport modified sleep stability and disrupted transitions between day and night sleep. In an Alzheimers disease model, metabolic interventions such as beta hydroxybutyrate supplementation and intermittent fasting selectively influenced the timing of day to night transitions rather than uniformly altering sleep duration. Autoencoder based similarity scoring and wavelet analysis reinforced that metabolic effects on sleep were highly time dependent. This study presents a novel approach to studying sleep-metabolism interactions, revealing that metabolic states exert their strongest influence at distinct time points, shaping sleep stability and circadian transitions.

q-bio.QM↗

arXiv2025

Sustainable Grid through Distributed Data Centers: Spinning AI Demand for Grid Stabilization and Optimization

We propose a disruptive paradigm to actively place and schedule TWhrs of parallel AI jobs strategically on the grid, at distributed, grid-aware high performance compute data centers (HPC) capable of using their massive power and energy load to stabilize the grid while reducing grid build-out requirements, maximizing use of renewable energy, and reducing Green House Gas (GHG) emissions. Our approach will enable the creation of new, value adding markets for spinning compute demand, providing market based incentives that will drive the joint optimization of energy and learning.

cs.DC↗

arXiv2025

Synchronized and Fine-Grained Head for Skeleton-Based Ambiguous Action Recognition

Skeleton-based action recognition using GCNs has achieved remarkable performance, but recognizing ambiguous actions, such as "waving" and "saluting", remains a significant challenge. Existing methods typically rely on a serial combination of GCNs and TCNs, where spatial and temporal features are extracted independently, leading to an unbalanced spatial-temporal information, which hinders accurate action recognition. Moreover, existing methods for ambiguous actions often overemphasize local details, resulting in the loss of crucial global context, which further complicates the task of differentiating ambiguous actions. To address these challenges, we propose a lightweight plug-and-play module called SF-Head, inserted between GCN and TCN layers. SF-Head first conducts SSTE with a Feature Redundancy Loss (F-RL), ensuring a balanced interaction. It then performs AC-FA, with a Feature Consistency Loss (F-CL), which aligns the aggregated feature with their original spatial-temporal feature. Experimental results on NTU RGB+D 60, NTU RGB+D 120, NW-UCLA and PKU-MMD I datasets demonstrate significant improvements in distinguishing ambiguous actions.

cs.CV↗

arXiv2025

TReND: Transformer derived features and Regularized NMF for neonatal functional network Delineation

Precise parcellation of functional networks (FNs) of early developing human brain is the fundamental basis for identifying biomarker of developmental disorders and understanding functional development. Resting-state fMRI (rs-fMRI) enables in vivo exploration of functional changes, but adult FN parcellations cannot be directly applied to the neonates due to incomplete network maturation. No standardized neonatal functional atlas is currently available. To solve this fundamental issue, we propose TReND, a novel and fully automated self-supervised transformer-autoencoder framework that integrates regularized nonnegative matrix factorization (RNMF) to unveil the FNs in neonates. TReND effectively disentangles spatiotemporal features in voxel-wise rs-fMRI data. The framework integrates confidence-adaptive masks into transformer self-attention layers to mitigate noise influence. A self supervised decoder acts as a regulator to refine the encoder's latent embeddings, which serve as reliable temporal features. For spatial coherence, we incorporate brain surface-based geodesic distances as spatial encodings along with functional connectivity from temporal features. The TReND clustering approach processes these features under sparsity and smoothness constraints, producing robust and biologically plausible parcellations. We extensively validated our TReND framework on three different rs-fMRI datasets: simulated, dHCP and HCP-YA against comparable traditional feature extraction and clustering techniques. Our results demonstrated the superiority of the TReND framework in the delineation of neonate FNs with significantly better spatial contiguity and functional homogeneity. Collectively, we established TReND, a novel and robust framework, for neonatal FN delineation. TReND-derived neonatal FNs could serve as a neonatal functional atlas for perinatal populations in health and disease.

q-bio.NC↗

arXiv2025

A Chain-of-Thought Subspace Meta-Learning for Few-shot Image Captioning with Large Vision and Language Models

A large-scale vision and language model that has been pretrained on massive data encodes visual and linguistic prior, which makes it easier to generate images and language that are more natural and realistic. Despite this, there is still a significant domain gap between the modalities of vision and language, especially when training data is scarce in few-shot settings, where only very limited data are available for training. In order to mitigate this issue, a multi-modal meta-learning framework has been proposed to bridge the gap between two frozen pretrained large vision and language models by introducing a tunable prompt connecting these two large models. For few-shot image captioning, the existing multi-model meta-learning framework utilizes a one-step prompting scheme to accumulate the visual features of input images to guide the language model, which struggles to generate accurate image descriptions with only a few training samples. Instead, we propose a chain-of-thought (CoT) meta-learning scheme as a multi-step image captioning procedure to better imitate how humans describe images. In addition, we further propose to learn different meta-parameters of the model corresponding to each CoT step in distinct subspaces to avoid interference. We evaluated our method on three commonly used image captioning datasets, i.e., MSCOCO, Flickr8k, and Flickr30k, under few-shot settings. The results of our experiments indicate that our chain-of-thought subspace meta-learning strategy is superior to the baselines in terms of performance across different datasets measured by different metrics.

cs.CV↗

arXiv2025

Integrating Retrospective Framework in Multi-Robot Collaboration

Recent advancements in Large Language Models (LLMs) have demonstrated substantial capabilities in enhancing communication and coordination in multi-robot systems. However, existing methods often struggle to achieve efficient collaboration and decision-making in dynamic and uncertain environments, which are common in real-world multi-robot scenarios. To address these challenges, we propose a novel retrospective actor-critic framework for multi-robot collaboration. This framework integrates two key components: (1) an actor that performs real-time decision-making based on observations and task directives, and (2) a critic that retrospectively evaluates the outcomes to provide feedback for continuous refinement, such that the proposed framework can adapt effectively to dynamic conditions. Extensive experiments conducted in simulated environments validate the effectiveness of our approach, demonstrating significant improvements in task performance and adaptability. This work offers a robust solution to persistent challenges in robotic collaboration.

cs.RO↗

arXiv2025

Note on hidden zeros and expansions of tree-level amplitudes

In this note, we derive and interpret hidden zeros of tree-level amplitudes of various theories, including Yang-Mills, non-linear sigma model, special Galileon, Dirac-Born-Infeld, and gravity, by utilizing universal expansions of tree-level amplitudes which expand amplitudes of different theories to those of bi-adjoint scalar theory. Hidden zeros of above amplitudes are attributed to zeros of bi-adjoint scalar amplitudes which can be easily proved. For unordered amplitudes such as gravitational ones, the kinematic condition for zeros causes potential divergences of propagators. We also show the detailed mechanism of eliminating such divergences.

hep-th↗

arXiv2025

RW-Net: Enhancing Few-Shot Point Cloud Classification with a Wavelet Transform Projection-based Network

In the domain of 3D object classification, a fundamental challenge lies in addressing the scarcity of labeled data, which limits the applicability of traditional data-intensive learning paradigms. This challenge is particularly pronounced in few-shot learning scenarios, where the objective is to achieve robust generalization from minimal annotated samples. To overcome these limitations, it is crucial to identify and leverage the most salient and discriminative features of 3D objects, thereby enhancing learning efficiency and reducing dependency on large-scale labeled datasets. This work introduces RW-Net, a novel framework designed to address the challenges above by integrating Rate-Distortion Explanation (RDE) and wavelet transform into a state-of-the-art projection-based 3D object classification architecture. The proposed method capitalizes on RDE to extract critical features by identifying and preserving the most informative data components while reducing redundancy. This process ensures the retention of essential information for effective decision-making, optimizing the model's ability to learn from limited data. Complementing RDE, incorporating the wavelet transform further enhances the framework's capability to generalize in low-data regimes. By emphasizing low-frequency components of the input data, the wavelet transform captures fundamental geometric and structural attributes of 3D objects. These attributes are instrumental in mitigating overfitting and improving the robustness of the learned representations across diverse tasks and domains. To validate the effectiveness of our RW-Net, we conduct extensive experiments on three datasets: ModelNet40, ModelNet40-C, and ScanObjectNN for few-shot 3D object classification. The results demonstrate that our approach achieves state-of-the-art performance and exhibits superior generalization and robustness in few-shot learning scenarios.

cs.CV↗

arXiv2024

An improved bound on Seymour's second neighborhood conjecture

Seymour's celebrated second neighborhood conjecture, now more than thirty years old, states that in every oriented digraph, there is a vertex $u$ such that the size of its second out-neighborhood $N^{++}(u)$ is at least as large as that of its first out-neighborhood $N^+(u)$. In this paper, we prove the existence of $u$ for which $|N^{++}(u)| \ge 0.715538 |N^+(u)|$. This result provides the first improvement to the best known constant factor in over two decades.

math.CO↗

arXiv2024

On the Temporal-spatial Analysis of Estimating Urban Traffic Patterns Via GPS Trace Data of Car-hailing Vehicles

Car-hailing services have become a prominent data source for urban traffic studies. Extracting useful information from car-hailing trace data is essential for effective traffic management, while discrepancies between car-hailing vehicles and urban traffic should be considered. This paper proposes a generic framework for estimating and analyzing urban traffic patterns using car-hailing trace data. The framework consists of three layers: the data layer, the interactive software layer, and the processing method layer. By pre-processing car-hailing GPS trace data with operations such as data cutting, map matching, and trace correction, the framework generates tensor matrices that estimate traffic patterns for car-hailing vehicle flow and average road speed. An analysis block based on these matrices examines the relationships and differences between car-hailing vehicles and urban traffic patterns, which have been overlooked in previous research. Experimental results demonstrate the effectiveness of the proposed framework in examining temporal-spatial patterns of car-hailing vehicles and urban traffic. For temporal analysis, urban road traffic displays a bimodal characteristic while car-hailing flow exhibits a 'multi-peak' pattern, fluctuating significantly during holidays and thus generating a hierarchical structure. For spatial analysis, the heat maps generated from the matrices exhibit certain discrepancies, but the spatial distribution of hotspots and vehicle aggregation areas remains similar.

stat.AP↗

arXiv2024

Neural-Network-Enhanced Metalens Camera for High-Definition, Dynamic Imaging in the Long-Wave Infrared Spectrum

To provide a lightweight and cost-effective solution for the long-wave infrared imaging using a singlet, we develop a camera by integrating a High-Frequency-Enhancing Cycle-GAN neural network into a metalens imaging system. The High-Frequency-Enhancing Cycle-GAN improves the quality of the original metalens images by addressing inherent frequency loss introduced by the metalens. In addition to the bidirectional cyclic generative adversarial network, it incorporates a high-frequency adversarial learning module. This module utilizes wavelet transform to extract high-frequency components, and then establishes a high-frequency feedback loop. It enables the generator to enhance the camera outputs by integrating adversarial feedback from the high-frequency discriminator. This ensures that the generator adheres to the constraints imposed by the high-frequency adversarial loss, thereby effectively recovering the camera's frequency loss. This recovery guarantees high-fidelity image output from the camera, facilitating smooth video production. Our camera is capable of achieving dynamic imaging at 125 frames per second with an End Point Error value of 12.58. We also achieve 0.42 for Fréchet Inception Distance, 30.62 for Peak Signal to Noise Ratio, and 0.69 for Structural Similarity in the recorded videos.

eess.IV↗

arXiv2024

Considerations and Recommendations from the ISMRM Diffusion Study Group for preclinical diffusion MRI: Part 1 -- In vivo small-animal imaging

Small-animal diffusion MRI (dMRI) has been used for methodological development and validation, characterizing the biological basis of diffusion phenomena, and comparative anatomy. The steps from animal setup and monitoring, to acquisition, analysis, and interpretation are complex, with many decisions that may ultimately affect what questions can be answered using the resultant data. This work aims to present selected recommendations and guidelines from the diffusion community, on best practices for preclinical dMRI of in vivo animals. We describe the general considerations and foundational knowledge that must be considered when designing experiments. We briefly describe differences in animal species and disease models and discuss why some may be more or less appropriate for different studies. We then give guidelines for in vivo acquisition protocols, including decisions on hardware, animal preparation, and imaging sequences, followed by advice for data processing including pre-processing, model-fitting, and tractography. Finally, we provide an online resource which lists publicly available preclinical dMRI datasets and software packages, to promote responsible and reproducible research. In each section, we attempt to provide guides and recommendations, but also highlight areas for which no guidelines exist (and why), and where future work should focus. While we mainly cover the central nervous system (on which most preclinical dMRI studies are focused), we also provide, where possible and applicable, recommendations for other organs of interest. An overarching goal herein is to enhance the rigor and reproducibility of small animal dMRI acquisitions and analyses, and thereby advance biomedical knowledge.

physics.med-ph↗

arXiv2024

A Review of Scalable and Privacy-Preserving Multi-Agent Frameworks for Distributed Energy Resources

Distributed energy resources (DERs) are gaining prominence due to their advantages in improving energy efficiency, reducing carbon emissions, and enhancing grid resilience. Despite the increasing deployment, the potential of DERs has yet to be fully explored and exploited. A fundamental question restrains the management of numerous DERs in large-scale power systems, "How should DER data be securely processed and DER operations be efficiently optimized?" To address this question, this paper considers two critical issues, namely privacy for processing DER data and scalability in optimizing DER operations, then surveys existing and emerging solutions from a multi-agent framework perspective. In the context of scalability, this paper reviews state-of-the-art research that relies on parallel control, optimization, and learning within distributed and/or decentralized information exchange structures, while in the context of privacy, it identifies privacy preservation measures that can be synthesized into the aforementioned scalable structures. Despite research advances in these areas, challenges remain because these highly interdisciplinary studies blend a wide variety of scalable computing architectures and privacy preservation techniques from different fields, making them difficult to adapt in practice. To mitigate this issue, this paper provides a holistic review of trending strategies that orchestrate privacy and scalability for large-scale power system operations from a multi-agent perspective, particularly for DER control problems. Furthermore, this review extrapolates new approaches for future scalable, privacy-aware, and cybersecure pathways to unlock the full potential of DERs through controlling, optimizing, and learning generic multi-agent-based cyber-physical systems.

eess.SY↗

arXiv2024

Benchmarking Complex Instruction-Following with Multiple Constraints Composition

Instruction following is one of the fundamental capabilities of large language models (LLMs). As the ability of LLMs is constantly improving, they have been increasingly applied to deal with complex human instructions in real-world scenarios. Therefore, how to evaluate the ability of complex instruction-following of LLMs has become a critical research problem. Existing benchmarks mainly focus on modeling different types of constraints in human instructions while neglecting the composition of different constraints, which is an indispensable constituent in complex instructions. To this end, we propose ComplexBench, a benchmark for comprehensively evaluating the ability of LLMs to follow complex instructions composed of multiple constraints. We propose a hierarchical taxonomy for complex instructions, including 4 constraint types, 19 constraint dimensions, and 4 composition types, and manually collect a high-quality dataset accordingly. To make the evaluation reliable, we augment LLM-based evaluators with rules to effectively verify whether generated texts can satisfy each constraint and composition. Furthermore, we obtain the final evaluation score based on the dependency structure determined by different composition types. ComplexBench identifies significant deficiencies in existing LLMs when dealing with complex instructions with multiple constraints composition.

cs.CL↗

arXiv2024

GAMap: Zero-Shot Object Goal Navigation with Multi-Scale Geometric-Affordance Guidance

Zero-Shot Object Goal Navigation (ZS-OGN) enables robots or agents to navigate toward objects of unseen categories without object-specific training. Traditional approaches often leverage categorical semantic information for navigation guidance, which struggles when only objects are partially observed or detailed and functional representations of the environment are lacking. To resolve the above two issues, we propose \textit{Geometric-part and Affordance Maps} (GAMap), a novel method that integrates object parts and affordance attributes as navigation guidance. Our method includes a multi-scale scoring approach to capture geometric-part and affordance attributes of objects at different scales. Comprehensive experiments conducted on HM3D and Gibson benchmark datasets demonstrate improvements in Success Rate and Success weighted by Path Length, underscoring the efficacy of our geometric-part and affordance-guided navigation approach in enhancing robot autonomy and versatility, without any additional object-specific training or fine-tuning with the semantics of unseen objects and/or the locomotions of the robot.

cs.RO↗

arXiv2024

Reliable Semantic Understanding for Real World Zero-shot Object Goal Navigation

We introduce an innovative approach to advancing semantic understanding in zero-shot object goal navigation (ZS-OGN), enhancing the autonomy of robots in unfamiliar environments. Traditional reliance on labeled data has been a limitation for robotic adaptability, which we address by employing a dual-component framework that integrates a GLIP Vision Language Model for initial detection and an InstructionBLIP model for validation. This combination not only refines object and environmental recognition but also fortifies the semantic interpretation, pivotal for navigational decision-making. Our method, rigorously tested in both simulated and real-world settings, exhibits marked improvements in navigation precision and reliability.

cs.RO↗