arXiv subjects

Jiawei Hu

Publications and source records attributed to Jiawei Hu.

At least 37 records · Page 2Linked to original sources

arXiv2025

Bayesian-based Online Label Shift Estimation with Dynamic Dirichlet Priors

Label shift, a prevalent challenge in supervised learning, arises when the class prior distribution of test data differs from that of training data, leading to significant degradation in classifier performance. To accurately estimate the test priors and enhance classification accuracy, we propose a Bayesian framework for label shift estimation, termed Full Maximum A Posterior Label Shift (FMAPLS), along with its online version, online-FMAPLS. Leveraging batch and online Expectation-Maximization (EM) algorithms, these methods jointly and dynamically optimize Dirichlet hyperparameters $\boldsymbolα$ and class priors $\boldsymbolπ$, thereby overcoming the rigid constraints of the existing Maximum A Posterior Label Shift (MAPLS) approach. Moreover, we introduce a linear surrogate function (LSF) to replace gradient-based hyperparameter updates, yielding closed-form solutions that reduce computational complexity while retaining asymptotic equivalence. The online variant substitutes the batch E-step with a stochastic approximation, enabling real-time adaptation to streaming data. Furthermore, our theoretical analysis reveals a fundamental trade-off between online convergence rate and estimation accuracy. Extensive experiments on CIFAR100 and ImageNet datasets under shuffled long-tail and Dirichlet test priors demonstrate that FMAPLS and online-FMAPLS respectively achieve up to 40% and 12% lower KL divergence and substantial improvements in post-shift accuracy over state-of-the-art baselines, particularly under severe class imbalance and distributional uncertainty. These results confirm the robustness, scalability, and suitability of the proposed methods for large-scale and dynamic learning scenarios.

cs.LG↗

arXiv2025

I&S-ViT: An Inclusive & Stable Method for Pushing the Limit of Post-Training ViTs Quantization

Albeit the scalable performance of vision transformers (ViTs), the dense computational costs (training & inference) undermine their position in industrial applications. Post-training quantization (PTQ), tuning ViTs with a tiny dataset and running in a low-bit format, well addresses the cost issue but unluckily bears more performance drops in lower-bit cases. In this paper, we introduce I&S-ViT, a novel method that regulates the PTQ of ViTs in an inclusive and stable fashion. I&S-ViT first identifies two issues in the PTQ of ViTs: (1) Quantization inefficiency in the prevalent log2 quantizer for post-Softmax activations; (2) Rugged and magnified loss landscape in coarse-grained quantization granularity for post-LayerNorm activations. Then, I&S-ViT addresses these issues by introducing: (1) A novel shift-uniform-log2 quantizer (SULQ) that incorporates a shift mechanism followed by uniform quantization to achieve both an inclusive domain representation and accurate distribution approximation; (2) A three-stage smooth optimization strategy (SOS) that amalgamates the strengths of channel-wise and layer-wise quantization to enable stable learning. Comprehensive evaluations across diverse vision tasks validate I&S-ViT' superiority over existing PTQ of ViTs methods, particularly in low-bit scenarios. For instance, I&S-ViT elevates the performance of 3-bit ViT-B by an impressive 50.68%.

cs.CV↗

arXiv2025

Molecular entanglement as a signature of the Unruh effect

The Unruh effect predicts that a uniformly accelerated observer perceives the vacuum seen by an inertial observer as a thermal bath at a temperature proportional to its proper acceleration. This phenomenon is often regarded as a flat spacetime ``cousin" of Hawking radiation. In this Letter, we first study the entanglement dynamics of a quantum system composed of two polarizable two-level subsystems undergoing centripetal acceleration in a vacuum. We demonstrate that the system's steady state can be entangled irrespective of the initial state, a distinct characteristic attributable to the circular manifestation of the Unruh effect. Through meticulous analysis, we then propose that this phenomenon can feasibly be detected using state-of-the-art optomechanical technologies, particularly with a quantum system of two molecules.

gr-qc↗

arXiv2025

Spontaneous excitation of a centripetally accelerated atom coupled to electromagnetic vacuum fluctuations near a reflecting boundary

We investigate the rate of change of the mean atomic energy for centripetally accelerated atoms interacting with electromagnetic vacuum fluctuations near a reflecting boundary, using the Dalibard-Dupont-Roc-Cohen-Tannoudji formalism. The distinct contributions from vacuum fluctuations and radiation reaction are analyzed separately. Our results reveal that, when the centripetal acceleration significantly exceeds the characteristic acceleration set by the atomic transition frequency, vacuum fluctuations dominates over radiation reaction, irrespective of the atom-boundary distance and the atomic polarization. In the near-zone regime, where the atom-boundary distance is much smaller than both the characteristic length associated with the acceleration and the transition wavelength of the atom, the boundary introduces substantial corrections to the rate of change of the mean atomic energy. These corrections are comparable in magnitude to those in free space and exhibit strong dependence on the atomic polarization. Remarkably, in the intermediate and far regions, contributions stemming from the combined effects of the boundary and acceleration can become the leading and subleading terms, respectively. An acceleration-independent term also arises from their interplay. These findings highlight the significant interplay between acceleration and the presence of a boundary in shaping atomic radiative properties and may have potential implications for experimentally probing the circular Unruh effect.

gr-qc↗

arXiv2025

Nonreciprocal RIS-Aided Covert Channel Reciprocity Attacks and Countermeasures

Reconfigurable intelligent surface (RIS) technology enhances wireless communication performance, but it also introduces new vulnerabilities that can be exploited by adversaries. This paper investigates channel reciprocity attack (CRACK) threats in multi-antenna wireless systems operating in time-division duplexing mode using a physically consistent non-reciprocal RIS (NR-RIS) model. CRACK can degrade communication rate and facilitate passive eavesdropping behavior by distorting the downlink precoding, without requiring any additional signal transmission or channel state information (CSI). Unlike conventional RIS jamming strategies, the NR-RIS does not need synchronization with the legitimate system and thus can operate with slow or fixed configurations to implement CRACK, obscuring the distinction between the direct and RIS-induced channels and thereby complicating corresponding defensive precoding designs. To counter the CRACK threat posed by NR-RIS, we develop ``SecureCoder,'' a deep reinforcement learning-based framework that can mitigate CRACK and determine an improved downlink precoder matrix using the estimated uplink CSI and rate feedback from the users. Simulation results demonstrate the severe performance degradation caused by NR-RIS CRACK and validate the effectiveness of SecureCoder in improving both throughput and reducing security threats, thereby enhancing system robustness.

eess.SP↗

arXiv2025

Unveiling the Landscape of Clinical Depression Assessment: From Behavioral Signatures to Psychiatric Reasoning

Depression is a widespread mental disorder that affects millions worldwide. While automated depression assessment shows promise, most studies rely on limited or non-clinically validated data, and often prioritize complex model design over real-world effectiveness. In this paper, we aim to unveil the landscape of clinical depression assessment. We introduce C-MIND, a clinical neuropsychiatric multimodal diagnosis dataset collected over two years from real hospital visits. Each participant completes three structured psychiatric tasks and receives a final diagnosis from expert clinicians, with informative audio, video, transcript, and functional near-infrared spectroscopy (fNIRS) signals recorded. Using C-MIND, we first analyze behavioral signatures relevant to diagnosis. We train a range of classical models to quantify how different tasks and modalities contribute to diagnostic performance, and dissect the effectiveness of their combinations. We then explore whether LLMs can perform psychiatric reasoning like clinicians and identify their clear limitations in realistic clinical settings. In response, we propose to guide the reasoning process with clinical expertise and consistently improves LLM diagnostic performance by up to 10% in Macro-F1 score. We aim to build an infrastructure for clinical depression assessment from both data and algorithmic perspectives, enabling C-MIND to facilitate grounded and reliable research for mental healthcare.

cs.CL↗

arXiv2025

CARTS: Cooperative and Adaptive Resource Triggering and Stitching for 5G ISAC

This paper presents CARTS, an adaptive 5G uplink sensing scheme designed to provide Integrated Sensing and Communication (ISAC) services. The performance of both communication and sensing fundamentally depends on the availability of accurate and up-to-date channel state information (CSI). In modern 5G networks, uplink CSI is derived from two reference signals: the demodulation reference signal (DMRS) and the sounding reference signal (SRS). However, current base station implementations treat these CSI measurements as separate information streams. The key innovation of CARTS is to fuse these two CSI streams, thereby increasing the frequency of CSI updates and extending sensing opportunities to more users. CARTS addresses two key challenges: (i) a novel channel stitching and compensation method that integrates asynchronous CSI estimates from DMRS and SRS, despite their different time and frequency allocations, and (ii) a real-time SRS triggering algorithm that complements the inherently uncontrollable DMRS schedule, ensuring sufficient and non-redundant sensing opportunities for all users. Our trace-driven evaluation shows that CARTS significantly improves scalability, achieving a channel estimation error (NMSE) of 0.167 and UE tracking accuracy of 85 cm while supporting twice the number of users as a periodic SRS-only baseline with similar performance. By opportunistically combining DMRS and SRS, CARTS therefore provides a practical, standard-compliant solution to improve CSI availability for ISAC without requiring additional radio resources.

cs.NI↗

arXiv2025

CheXPO: Preference Optimization for Chest X-ray VLMs with Counterfactual Rationale

Vision-language models (VLMs) are prone to hallucinations that critically compromise reliability in medical applications. While preference optimization can mitigate these hallucinations through clinical feedback, its implementation faces challenges such as clinically irrelevant training samples, imbalanced data distributions, and prohibitive expert annotation costs. To address these challenges, we introduce CheXPO, a Chest X-ray Preference Optimization strategy that combines confidence-similarity joint mining with counterfactual rationale. Our approach begins by synthesizing a unified, fine-grained multi-task chest X-ray visual instruction dataset across different question types for supervised fine-tuning (SFT). We then identify hard examples through token-level confidence analysis of SFT failures and use similarity-based retrieval to expand hard examples for balancing preference sample distributions, while synthetic counterfactual rationales provide fine-grained clinical preferences, eliminating the need for additional expert input. Experiments show that CheXPO achieves 8.93% relative performance gain using only 5% of SFT samples, reaching state-of-the-art performance across diverse clinical tasks and providing a scalable, interpretable solution for real-world radiology applications.

cs.CV↗

arXiv2025

YOLO-FDA: Integrating Hierarchical Attention and Detail Enhancement for Surface Defect Detection

Surface defect detection in industrial scenarios is both crucial and technically demanding due to the wide variability in defect types, irregular shapes and sizes, fine-grained requirements, and complex material textures. Although recent advances in AI-based detectors have improved performance, existing methods often suffer from redundant features, limited detail sensitivity, and weak robustness under multiscale conditions. To address these challenges, we propose YOLO-FDA, a novel YOLO-based detection framework that integrates fine-grained detail enhancement and attention-guided feature fusion. Specifically, we adopt a BiFPN-style architecture to strengthen bidirectional multilevel feature aggregation within the YOLOv5 backbone. To better capture fine structural changes, we introduce a Detail-directional Fusion Module (DDFM) that introduces a directional asymmetric convolution in the second-lowest layer to enrich spatial details and fuses the second-lowest layer with low-level features to enhance semantic consistency. Furthermore, we propose two novel attention-based fusion strategies, Attention-weighted Concatenation (AC) and Cross-layer Attention Fusion (CAF) to improve contextual representation and reduce feature noise. Extensive experiments on benchmark datasets demonstrate that YOLO-FDA consistently outperforms existing state-of-the-art methods in terms of both accuracy and robustness across diverse types of defects and scales.

cs.CV↗

arXiv2025

Solar-CSK: Decoding Color Coded Visible Light Communications using Solar Cells

Visible Light Communication (VLC) provides an energy-efficient wireless solution by using existing LED-based illumination for high-speed data transmissions. Although solar cells offer the advantage of simultaneous energy harvesting and data reception, their broadband nature hinders accurate decoding of color-coded signals like Color Shift Keying (CSK). In this paper, we propose a novel approach exploiting the concept of tandem solar cells, multi-layer devices with partial wavelength selectivity, to capture coarse color information without resorting to energy-limiting color filters. To address the residual spectral overlap, we develop a bidirectional LSTM-based machine learning framework that infers channel characteristics by comparing solar cells' photovoltaic signals with pilot-based anchor data. Our commercial off-the-shelf (COTS) solar prototype achieves robust performance across varying distances and ambient lighting levels, significantly reducing bit error rates compared to conventional channel estimation methods. These findings mark a step toward sustainable, high-performance VLC systems powered by the multi-layer solar technologies.

cs.NI↗

arXiv2025

Gravitational lensing by charged black hole with global monopole in the strong field limit

We investigate gravitational lensing near a charged black hole with a global monopole in the strong field regime, focusing on the combined effects of the global monopole and black hole charge on key observables in gravitational lensing both analytically and numerically. Our results reveal that the dependence of the angular separation on charge is intricately tied to the deficit angle caused by the global monopole. In particular, we identify three critical values of the global monopole parameter that determine whether the angular separation increases monotonically, decreases monotonically, or exhibits extrema as the charge varies. A similar complex dependence is found for the flux ratio as a function of the deficit angle, and for the magnification of the first relativistic image as a function of charge. These behaviors contrast sharply with the monotonic changes observed in the absence of either a global monopole or charge. Our findings highlight that the effects of the charge and global monopole on gravitational lensing cannot be described as simple additive contributions. Instead, their combined effects lead to a rich and interdependent behavior that enhances our understanding of strong-field gravitational lensing. While the charge and global monopole are expected to be small in typical astrophysical contexts, the results presented here could be experimentally explored in analog gravity systems, where these parameters are not constrained. This opens the door to potential experimental verification of the phenomena predicted in this study.

gr-qc↗

arXiv2025

Significant circular Unruh effect at small acceleration

We study the transition rates of an atom rotating in a circular orbit, which is coupled with fluctuating electromagnetic fields in vacuum. We find that when the rotational angular velocity exceeds the transition frequency of the atom, the excitation rate can reach the same order of magnitude as the emission rate, even with an extremely low centripetal acceleration resulting from a very small orbital radius. For experimentally accessible centripetal accelerations, the excitation rate of centripetally accelerated atoms can be $10^{272,878}$ times greater than that of linearly accelerated atoms with the same magnitude of acceleration. Our result suggests that the circular version of the Unruh effect can be significant even at very small centripetal accelerations, contrary to the common belief that a large Unruh effect requires large acceleration. This finding sheds new light on the experimental detection of the circular Unruh effect.

gr-qc↗

arXiv2025

Electrothermal manipulation of current-induced phase transitions in ferrimagnetic Mn$_3$Si$_2$Te$_6$

Phase transitions driven by external stimuli are central to condensed matter physics, providing critical insights into symmetry breaking and emergent phenomena. Recently, ferrimagnetic (FiM) Mn$_3$Si$_2$Te$_6$ has attracted considerable attention for its magnetic-field-induced insulator-metal transitions (IMTs) and unconventional current-driven phase transitions, yet the role of applied currents in the magnetic phase remains poorly understood. Here, by combining local magnetization probes and time-resolved transport measurements, we uncover an electrothermal origin for the current-induced first-order-like phase transitions, characterized by abrupt voltage jumps and distinct magnetic domain evolution. Current-voltage (I-V) characteristics measured under triangular waveforms exhibit strong non-reciprocal and hysteretic behaviors, which are significantly suppressed at frequencies ~1000 Hz. Time-resolved studies using rectangular pulsed currents demonstrate that the resistance dynamics closely mirror the equilibrium resistance-temperature profile, directly implicating Joule heating as the driving mechanism. Furthermore, we reveal that the intrinsic I-V response adheres to Ohm's law, displaying linearity across various magnetic fields and temperatures. Our work advocates for a cautious approach in distinguishing between genuine current-induced nonequilibrium quantum states and thermal effects.

cond-mat.mtrl-sci↗

arXiv2025

Repulsive quantum gravitoelectric-gravitomagnetic interaction

We investigate, in the framework of linearized quantum gravity, the quantum gravitational interaction between a gravitoelectrically polarizable object and a gravitomagnetically polarizable object. This interaction originates from the coupling between the instantaneous mass quadrupole moment and the mass-current quadrupole moment of the objects, induced by fluctuating gravitoelectric and gravitomagnetic fields in a vacuum. Using leading-order perturbation theory, we derive the explicit expression of the quantum gravitoelectric-gravitomagnetic interaction energy, which shows a distance dependence of $r^{-8}$ in the near regime and $r^{-11}$ in the far regime, where $r$ is the distance between the two objects. Remarkably, this interaction energy is positive, indicating that the force is repulsive. Since interactions between objects polarizable in the same gravitoelectric or gravitomagnetic manner are inherently attractive, for objects which are both gravitoelectrically and gravitomagnetically polarizable, the overall quantum gravitational interaction potential is reduced when the repulsive quantum gravitoelectric-gravitomagnetic interaction is taken into account. However, for two isotropically polarizable objects with identical gravitoelectric and gravitomagnetic polarizabilities and energy level spacing, the repulsive quantum interaction cannot surpass the attractive interactions.

gr-qc↗

arXiv2025

Towards Accurate Post-Training Quantization of Vision Transformers via Error Reduction

Post-training quantization (PTQ) for vision transformers (ViTs) has received increasing attention from both academic and industrial communities due to its minimal data needs and high time efficiency. However, many current methods fail to account for the complex interactions between quantized weights and activations, resulting in significant quantization errors and suboptimal performance. This paper presents ERQ, an innovative two-step PTQ method specifically crafted to reduce quantization errors arising from activation and weight quantization sequentially. The first step, Activation quantization error reduction (Aqer), first applies Reparameterization Initialization aimed at mitigating initial quantization errors in high-variance activations. Then, it further mitigates the errors by formulating a Ridge Regression problem, which updates the weights maintained at full-precision using a closed-form solution. The second step, Weight quantization error reduction (Wqer), first applies Dual Uniform Quantization to handle weights with numerous outliers, which arise from adjustments made during Reparameterization Initialization, thereby reducing initial weight quantization errors. Then, it employs an iterative approach to further tackle the errors. In each iteration, it adopts Rounding Refinement that uses an empirically derived, efficient proxy to refine the rounding directions of quantized weights, complemented by a Ridge Regression solver to reduce the errors. Comprehensive experimental results demonstrate ERQ's superior performance across various ViTs variants and tasks. For example, ERQ surpasses the state-of-the-art GPTQ by a notable 36.81% in accuracy for W3A4 ViT-S. Our codes are available at https://github.com/zysxmu/ERQ.

cs.CV↗

arXiv2024

Mirror Construction for Nakajima Quiver Varieties

In this paper, we construct the ADHM quiver representations and the corresponding sheaves as the mirror objects of formal deformations of the framed immersed Lagrangian sphere decorated with flat bundles. More generally, we construct Nakajima quiver varieties as localized mirrors of framed nodal unions of Lagrangian spheres in dimension two. This produces a mirror functor from the Fukaya category of a framed plumbing of surfaces to the dg category of complexes of bundles over the corresponding Nakajima quiver varieties. For affine ADE quivers in specific multiplicities, the corresponding (unframed) Lagrangian immersions are homological tori, whose moduli of stable deformations are asymptotically locally Euclidean (ALE) spaces. We show that framed stable Lagrangian branes are transformed into monadic complexes of framed torsion-free sheaves over the ALE spaces. A main ingredient is the notion of framed Lagrangian immersions and their Maurer-Cartan deformations. Moreover, using the formalism of quiver algebroid stacks, we find isomorphisms between the moduli of stable Lagrangian immersions and that of special Lagrangian fibers of an SYZ fibration in the affine $A_n$ cases.

math.AG↗

arXiv2024

LightLLM: A Versatile Large Language Model for Predictive Light Sensing

We propose LightLLM, a model that fine tunes pre-trained large language models (LLMs) for light-based sensing tasks. It integrates a sensor data encoder to extract key features, a contextual prompt to provide environmental information, and a fusion layer to combine these inputs into a unified representation. This combined input is then processed by the pre-trained LLM, which remains frozen while being fine-tuned through the addition of lightweight, trainable components, allowing the model to adapt to new tasks without altering its original parameters. This approach enables flexible adaptation of LLM to specialized light sensing tasks with minimal computational overhead and retraining effort. We have implemented LightLLM for three light sensing tasks: light-based localization, outdoor solar forecasting, and indoor solar estimation. Using real-world experimental datasets, we demonstrate that LightLLM significantly outperforms state-of-the-art methods, achieving 4.4x improvement in localization accuracy and 3.4x improvement in indoor solar estimation when tested in previously unseen environments. We further demonstrate that LightLLM outperforms ChatGPT-4 with direct prompting, highlighting the advantages of LightLLM's specialized architecture for sensor data fusion with textual prompts.

cs.LG↗

arXiv2024

Quantum gravitomagnetic interaction

In the framework of linearized quantum gravity, we study the quantum gravitational interaction between two nonpointlike objects induced by fluctuating gravitomagnetic fields in vacuum. We find that, in addition to the quantum gravitational interaction induced by fluctuating gravitoelectric fields previously studied, there exists a quantum gravitomagnetic interaction. This interaction originates from the interaction between the instantaneous localized mass currents in nonpointlike objects induced by the fluctuating gravitomagnetic fields. Using fourth-order perturbation theory, we derive the explicit form of the quantum gravitomagnetic interaction energy, which shows an $r^{-10}$ dependence in the near regime and an $r^{-11}$ dependence in the far regime, where $r$ is the distance between the two objects. This interaction energy is expected to be significant when the gravitomagnetic polarizability of the objects is large.

gr-qc↗