arXiv subjects

Hui Chen

Publications and source records attributed to Hui Chen.

At least 127 records · Page 7Linked to original sources

arXiv2025

Cross-Lingual Consistency: A Novel Inference Framework for Advancing Reasoning in Large Language Models

Chain-of-thought (CoT) has emerged as a critical mechanism for enhancing reasoning capabilities in large language models (LLMs), with self-consistency demonstrating notable promise in boosting performance. However, inherent linguistic biases in multilingual training corpora frequently cause semantic drift and logical inconsistencies, especially in sub-10B parameter LLMs handling complex inference tasks. To overcome these constraints, we propose the Cross-Lingual Consistency (CLC) framework, an innovative inference paradigm that integrates multilingual reasoning paths through majority voting to elevate LLMs' reasoning capabilities. Empirical evaluations on the CMATH dataset reveal CLC's superiority over the conventional self-consistency method, delivering 9.5%, 6.5%, and 6.0% absolute accuracy gains for DeepSeek-Math-7B-Instruct, Qwen2.5-Math-7B-Instruct, and Gemma2-9B-Instruct respectively. Expanding CLC's linguistic scope to 11 diverse languages implies two synergistic benefits: 1) neutralizing linguistic biases in multilingual training corpora through multilingual ensemble voting, 2) escaping monolingual reasoning traps by exploring the broader multilingual solution space. This dual benefits empirically enables more globally optimal reasoning paths compared to monolingual self-consistency baselines, as evidenced by the 4.1%-18.5% accuracy gains using Gemma2-9B-Instruct on the MGSM dataset.

cs.CL↗

arXiv2025

LSNet: See Large, Focus Small

Vision network designs, including Convolutional Neural Networks and Vision Transformers, have significantly advanced the field of computer vision. Yet, their complex computations pose challenges for practical deployments, particularly in real-time applications. To tackle this issue, researchers have explored various lightweight and efficient network designs. However, existing lightweight models predominantly leverage self-attention mechanisms and convolutions for token mixing. This dependence brings limitations in effectiveness and efficiency in the perception and aggregation processes of lightweight networks, hindering the balance between performance and efficiency under limited computational budgets. In this paper, we draw inspiration from the dynamic heteroscale vision ability inherent in the efficient human vision system and propose a ``See Large, Focus Small'' strategy for lightweight vision network design. We introduce LS (\textbf{L}arge-\textbf{S}mall) convolution, which combines large-kernel perception and small-kernel aggregation. It can efficiently capture a wide range of perceptual information and achieve precise feature aggregation for dynamic and complex visual representations, thus enabling proficient processing of visual information. Based on LS convolution, we present LSNet, a new family of lightweight models. Extensive experiments demonstrate that LSNet achieves superior performance and efficiency over existing lightweight networks in various vision tasks. Codes and models are available at https://github.com/jameslahm/lsnet.

cs.CV↗

arXiv2025

DiffusionTalker: Efficient and Compact Speech-Driven 3D Talking Head via Personalizer-Guided Distillation

Real-time speech-driven 3D facial animation has been attractive in academia and industry. Traditional methods mainly focus on learning a deterministic mapping from speech to animation. Recent approaches start to consider the nondeterministic fact of speech-driven 3D face animation and employ the diffusion model for the task. Existing diffusion-based methods can improve the diversity of facial animation. However, personalized speaking styles conveying accurate lip language is still lacking, besides, efficiency and compactness still need to be improved. In this work, we propose DiffusionTalker to address the above limitations via personalizer-guided distillation. In terms of personalization, we introduce a contrastive personalizer that learns identity and emotion embeddings to capture speaking styles from audio. We further propose a personalizer enhancer during distillation to enhance the influence of embeddings on facial animation. For efficiency, we use iterative distillation to reduce the steps required for animation generation and achieve more than 8x speedup in inference. To achieve compactness, we distill the large teacher model into a smaller student model, reducing our model's storage by 86.4\% while minimizing performance loss. After distillation, users can derive their identity and emotion embeddings from audio to quickly create personalized animations that reflect specific speaking styles. Extensive experiments are conducted to demonstrate that our method outperforms state-of-the-art methods. The code will be released at: https://github.com/ChenVoid/DiffusionTalker.

cs.CV↗

arXiv2025

Cream of the Crop: Harvesting Rich, Scalable and Transferable Multi-Modal Data for Instruction Fine-Tuning

The hypothesis that pretrained large language models (LLMs) necessitate only minimal supervision during the fine-tuning (SFT) stage (Zhou et al., 2024) has been substantiated by recent advancements in data curation and selection research. However, their stability and generalizability are compromised due to the vulnerability to experimental setups and validation protocols, falling short of surpassing random sampling (Diddee & Ippolito, 2024; Xia et al., 2024b). Built upon LLMs, multi-modal LLMs (MLLMs), combined with the sheer token volume and heightened heterogeneity of data sources, amplify both the significance and complexity of data selection. To harvest multi-modal instructional data in a robust and efficient manner, we re-define the granularity of the quality metric by decomposing it into 14 vision-language-related capabilities, and introduce multi-modal rich scorers to evaluate the capabilities of each data candidate. To promote diversity, in light of the inherent objective of the alignment stage, we take interaction style as diversity indicator and use a multi-modal rich styler to identify data instruction patterns. In doing so, our multi-modal rich scorers and styler (mmSSR) guarantee that high-scoring information is conveyed to users in diversified forms. Free from embedding-based clustering or greedy sampling, mmSSR efficiently scales to millions of data with varying budget constraints, supports customization for general or specific capability acquisition, and facilitates training-free generalization to new domains for curation. Across 10+ experimental settings, validated by 14 multi-modal benchmarks, we demonstrate consistent improvements over random sampling, baseline strategies and state-of-the-art selection methods, achieving 99.1% of full performance with only 30% of the 2.6M data.

cs.CV↗

arXiv2025

Discovery of unconventional charge-spin-intertwined density wave in magnetic kagome metal GdTi3Bi4

The symmetry breaking and its interplay among spin, charge, and lattice degrees of freedom is crucial for understanding correlated quantum states such as charge density waves (CDWs) and unconventional superconductivity. Here, we report the discovery by low-temperature scanning tunneling microscopy/spectroscopy of unconventional charge-spin-intertwined density waves in magnetic kagome metal GdTi3Bi4, which exhibits the one-third magnetization plateau. We reveal the emergence of 3Q CDWs incommensurate with the crystalline lattice in both periodicity and orientation, breaking all mirror and rotation symmetries. The CDW exhibits incommensurate-commensurate transitions in an applied magnetic field and transitions between 3Q and 1Q CDWs as a function of field and temperature, accompanied by changes in the spatial symmetries. Remarkably, the quantum and classic melting of the CDWs exhibits a phase structure which is consistent with the magnetization phase diagram of bulk GdTi3Bi4, providing strong evidence for the intertwined charge-spin density wave order. The origin of the charge-spin intertwinement is further evidenced by the observed hybridization between itinerant electrons and Gd local moments. Our findings uncover an unconventional form of charge-spin orders and offer new insights into a broad class of multi-components density wave formation in kagome and other correlated quantum materials.

cond-mat.str-el↗

arXiv2025

Joint Bistatic Positioning and Monostatic Sensing: Optimized Beamforming and Performance Tradeoff

We investigate joint bistatic positioning (BP) and monostatic sensing (MS) within a multi-input multi-output orthogonal frequency-division system. Based on the derived Cramér-Rao Bounds (CRBs), we propose novel beamforming optimization strategies that enable flexible performance trade-offs between BP and MS. Two distinct objectives are considered in this multi-objective optimization problem, namely, enabling user equipment to estimate its own position while accounting for unknown clock bias and orientation, and allowing the base station to locate passive targets. We first analyze digital schemes, proposing both weighted-sum CRB and weighted-sum mismatch (of beamformers and covariance matrices) minimization approaches. These are examined under full-dimension beamforming (FDB) and low-complexity codebook-based power allocation (CPA). To adapt to low-cost hardwares, we develop unit-amplitude analog FDB and CPA schemes based on the weighted-sum mismatch of the covariance matrices paradigm, solved using distinct methods. Numerical results confirm the effectiveness of our designs, highlighting the superiority of minimizing the weighted-sum mismatch of covariance matrices, and the advantages of mutual information fusion between BP and MS.

eess.SP↗

arXiv2025

Privacy Preservation in MIMO-OFDM Localization Systems: A Beamforming Approach

We investigate an uplink MIMO-OFDM localization scenario where a legitimate base station (BS) aims to localize a user equipment (UE) using pilot signals transmitted by the UE, while an unauthorized BS attempts to localize the UE by eavesdropping on these pilots, posing a risk to the UE's location privacy. To enhance legitimate localization performance while protecting the UE's privacy, we formulate an optimization problem regarding the beamformers at the UE, aiming to minimize the Cramér-Rao bound (CRB) for legitimate localization while constraining the CRB for unauthorized localization above a threshold. A penalty dual decomposition optimization framework is employed to solve the problem, leading to a novel beamforming approach for location privacy preservation. Numerical results confirm the effectiveness of the proposed approach and demonstrate its superiority over existing benchmarks.

eess.SP↗

arXiv2025

Optimized Beamforming for Joint Bistatic Positioning and Monostatic Sensing

We investigate the performance tradeoff between \textit{bistatic positioning (BP)} and \textit{monostatic sensing (MS)} in a multi-input multi-output orthogonal frequency division multiplexing scenario. We derive the Cramér-Rao bounds (CRBs) for BP at the user equipment and MS at the base station. To balance these objectives, we propose a multi-objective optimization framework that optimizes beamformers using a weighted-sum CRB approach, ensuring the weak Pareto boundary. We also introduce two mismatch-minimizing approaches, targeting beamformer mismatch and variance matrix mismatch, and solve them distinctly. Numerical results demonstrate the performance tradeoff between BP and MS, revealing significant gains with the proposed methods and highlighting the advantages of minimizing the weighted-sum mismatch of variance matrices.

eess.SP↗

arXiv2025

A Dynamic Model of Private Asset Allocation

We build a state-of-the-art dynamic model of private asset allocation that considers five key features of private asset markets: (1) the illiquid nature of private assets, (2) timing lags between capital commitments, capital calls, and eventual distributions, (3) time-varying business cycle conditions, (4) serial correlation in observed private asset returns, and (5) regulatory constraints on certain institutional investors' portfolio choices. We use cutting-edge machine learning methods to quantify the optimal investment policies over the life cycle of a fund. Moreover, our model offers regulators a tool for precisely quantifying the trade-offs when setting risk-based capital charges.

q-fin.PM↗

arXiv2025

Joint Near-Field Sensing and Visibility Region Detection with Extremely Large Aperture Arrays

In this paper, we consider near-field localization and sensing with an extremely large aperture array under partial blockage of array antennas, where spherical wavefront and spatial non-stationarity are accounted for. We propose an Ising model to characterize the clustered sparsity feature of the blockage pattern, develop an algorithm based on alternating optimization for joint channel parameter estimation and visibility region detection, and further estimate the locations of the user and environmental scatterers. The simulation results confirm the effectiveness of the proposed algorithm compared to conventional methods.

eess.SP↗

arXiv2025

RIS-Aided Positioning Under Adverse Conditions: Interference from Unauthorized RIS

Positioning technology, which aims to determine the geometric information of a device in a global coordinate, is a key component in integrated sensing and communication systems. In addition to traditional active anchor-based positioning systems, reconfigurable intelligent surfaces (RIS) have shown great potential for enhancing system performance. However, their ability to manipulate electromagnetic waves and ease of deployment pose potential risks, as unauthorized RIS may be intentionally introduced to jeopardize the positioning service. Such an unauthorized RIS can cause unexpected interference in the original localization system, distorting the transmitted signals, and leading to degraded positioning accuracy. In this work, we investigate the scenario of RIS-aided positioning in the presence of interference from an unauthorized RIS. Theoretical lower bounds are employed to analyze the impact of unauthorized RIS on channel parameter estimation and positioning accuracy. Several codebook design strategies for unauthorized RIS are evaluated, and various system arrangements are discussed. The simulation results show that an unauthorized RIS path with a high channel gain or a delay similar to that of legitimate RIS paths leads to poor positioning performance. Furthermore, unauthorized RIS generates more effective interference when using directional beamforming codebooks compared to random codebooks.

eess.SP↗

arXiv2025

Intrinsic Cramér-Rao Bound based 6D Localization and Tracking for 5G/6G Systems

Localization and tracking are critical components of integrated sensing and communication (ISAC) systems, enhancing resource management, beamforming accuracy, and overall system reliability through precise sensing. Due to the high path loss of the high-frequency systems, antenna arrays are required at the transmitter and receiver sides for beamforming gain. However, beam misalignment may occur, which requires accurate tracking of the six-dimensional (6D) state, namely, 3D position and 3D orientation. In this work, we first address the challenge that the rotation matrix, being part of the Lie group rather than Euclidean space, necessitates the derivation of the ICRB for an intrinsic performance benchmark. Then, leveraging the derived ICRB, we develop two filters-one utilizing pose fusion and the other employing error-state Kalman filter to estimate the UE's 6D state for different computational resource consumption and accuracy requirements. Simulation results validate the ICRB and assess the performance of the proposed filters, demonstrating their effectiveness and improved accuracy in 6D state tracking.

eess.SP↗

arXiv2025

CartesianMoE: Boosting Knowledge Sharing among Experts via Cartesian Product Routing in Mixture-of-Experts

Large language models (LLM) have been attracting much attention from the community recently, due to their remarkable performance in all kinds of downstream tasks. According to the well-known scaling law, scaling up a dense LLM enhances its capabilities, but also significantly increases the computational complexity. Mixture-of-Experts (MoE) models address that by allowing the model size to grow without substantially raising training or inference costs. Yet MoE models face challenges regarding knowledge sharing among experts, making their performance somehow sensitive to routing accuracy. To tackle that, previous works introduced shared experts and combined their outputs with those of the top $K$ routed experts in an ``addition'' manner. In this paper, inspired by collective matrix factorization to learn shared knowledge among data, we propose CartesianMoE, which implements more effective knowledge sharing among experts in more like a ``multiplication'' manner. Extensive experimental results indicate that CartesianMoE outperforms previous MoE models for building LLMs, in terms of both perplexity and downstream task performance. And we also find that CartesianMoE achieves better expert routing robustness.

cs.LG↗

arXiv2025

Finedeep: Mitigating Sparse Activation in Dense LLMs via Multi-Layer Fine-Grained Experts

Large language models have demonstrated exceptional performance across a wide range of tasks. However, dense models usually suffer from sparse activation, where many activation values tend towards zero (i.e., being inactivated). We argue that this could restrict the efficient exploration of model representation space. To mitigate this issue, we propose Finedeep, a deep-layered fine-grained expert architecture for dense models. Our framework partitions the feed-forward neural network layers of traditional dense models into small experts, arranges them across multiple sub-layers. A novel routing mechanism is proposed to determine each expert's contribution. We conduct extensive experiments across various model sizes, demonstrating that our approach significantly outperforms traditional dense architectures in terms of perplexity and benchmark performance while maintaining a comparable number of parameters and floating-point operations. Moreover, we find that Finedeep achieves optimal results when balancing depth and width, specifically by adjusting the number of expert sub-layers and the number of experts per sub-layer. Empirical results confirm that Finedeep effectively alleviates sparse activation and efficiently utilizes representation capacity in dense models.

cs.CL↗

arXiv2025

LLMI3D: MLLM-based 3D Perception from a Single 2D Image

Recent advancements in autonomous driving, augmented reality, robotics, and embodied intelligence have necessitated 3D perception algorithms. However, current 3D perception methods, especially specialized small models, exhibit poor generalization in open scenarios. On the other hand, multimodal large language models (MLLMs) excel in general capacity but underperform in 3D tasks, due to weak 3D local spatial object perception, poor text-based geometric numerical output, and inability to handle camera focal variations. To address these challenges, we propose the following solutions: Spatial-Enhanced Local Feature Mining for better spatial feature extraction, 3D Query Token-Derived Info Decoding for precise geometric regression, and Geometry Projection-Based 3D Reasoning for handling camera focal length variations. We employ parameter-efficient fine-tuning for a pre-trained MLLM and develop LLMI3D, a powerful 3D perception MLLM. Additionally, we have constructed the IG3D dataset, which provides fine-grained descriptions and question-answer annotations. Extensive experiments demonstrate that our LLMI3D achieves state-of-the-art performance, outperforming other methods by a large margin.

cs.CV↗

arXiv2025

Revealing the orbital origins of exotic electronic states with Ti substitution in kagome superconductor CsV3Sb5

The multiband kagome superconductor CsV3Sb5 exhibits complex orbital textures on the Fermi surface, making the orbital origins of its cascade of correlated electronic states and superconductivity a major scientific puzzle. Chemical doping of the kagome plane can simultaneously tune the exotic states and the Fermi-surface orbital texture, and thus offers a unique opportunity to correlate the given states with specific orbitals. In this Letter, by substituting V atoms with Ti in kagome superconductor CsV3Sb5, we reveal the orbital origin of a cascade of its correlated electronic states through the orbital-resolved quasiparticle interference (QPI). We analyze the QPI changes associated with different orbitals, aided by first-principles calculations. We have observed that the in-plane and out-of-plane vanadium 3d orbitals cooperate to form unidirectional coherent states in pristine CsV3Sb5, whereas the out-of-plane component disappears with doping-induced suppression of charge density wave and global electronic nematicity. In addition, the Sb pz orbital plays an important role in both the pseudo-gap and superconducting states in CsV3Sb5. Our findings offer new insights into multiorbital physics in quantum materials which are generally manifested with intriguing correlations between atomic orbitals and symmetry-encoded correlated electronic states.

cond-mat.supr-con↗

arXiv2025

Integrated 6G TN and NTN Localization: Challenges, Opportunities, and Advancements

The rapid evolution of cellular networks has introduced groundbreaking technologies, including large and distributed antenna arrays and reconfigurable intelligent surfaces in terrestrial networks (TNs), as well as aerial and space-based nodes in non-terrestrial networks (NTNs). These advancements enable applications beyond traditional communication, such as high-precision localization and sensing. While integrating TN and NTN enablers will lead to unparalleled opportunities for seamless global localization, such integration attempts are expected to face several challenges. To understand these opportunities and challenges, we first examine the distinctive characteristics of the key 6G enablers, evaluating their roles in localization from both technical and practical perspectives. Next, to identify developments driving TN-NTN localization, we review the latest standardization and industrial innovation progress. Finally, we discuss the opportunities and challenges of TN-NTN integration, illustrating its potential through two numerical case studies.

eess.SP↗

arXiv2025

Federated Neural Nonparametric Point Processes

Temporal point processes (TPPs) are effective for modeling event occurrences over time, but they struggle with sparse and uncertain events in federated systems, where privacy is a major concern. To address this, we propose \textit{FedPP}, a Federated neural nonparametric Point Process model. FedPP integrates neural embeddings into Sigmoidal Gaussian Cox Processes (SGCPs) on the client side, which is a flexible and expressive class of TPPs, allowing it to generate highly flexible intensity functions that capture client-specific event dynamics and uncertainties while efficiently summarizing historical records. For global aggregation, FedPP introduces a divergence-based mechanism that communicates the distributions of SGCPs' kernel hyperparameters between the server and clients, while keeping client-specific parameters local to ensure privacy and personalization. FedPP effectively captures event uncertainty and sparsity, and extensive experiments demonstrate its superior performance in federated settings, particularly with KL divergence and Wasserstein distance-based global aggregation.

cs.LG↗