arXiv subjects

Qiang Wu

Publications and source records attributed to Qiang Wu.

At least 55 records · Page 3Linked to original sources

arXiv2025

ASVD: Activation-aware Singular Value Decomposition for Compressing Large Language Models

In this paper, we introduce a new post-training compression paradigm for Large Language Models (LLMs) to facilitate their wider adoption. We delve into LLM weight low-rank decomposition, and find that the challenges of this task stem from (1) the distribution variance in the LLM activations and (2) the sensitivity difference among various kinds of layers. To address these issues, we propose a training-free approach called Activation-aware Singular Value Decomposition (ASVD). Specifically, ASVD manages activation outliers by transforming the weight matrix based on the activation distribution. This transformation allows the outliers in the activation matrix to be absorbed into the transformed weight matrix, thereby enhancing decomposition accuracy. Additionally, we propose an efficient iterative calibration process to optimize layer-specific decomposition by addressing the varying sensitivity of different LLM layers. In this way, ASVD can compress a network by 10%-30%. Based on the success of the low-rank decomposition of projection matrices in the self-attention module, we further introduce ASVD to compress the KV cache. By reducing the channel dimension of KV activations, memory requirements for KV cache can be largely reduced. ASVD can further achieve 50% KV cache reductions without performance drop in a training-free manner.

cs.CL↗

arXiv2025

Lyapunov Exponent Approach to Phase Structure of Schwarzschild AdS Black Holes Surrounded by a Cloud of Strings

We investigate Schwarzschild black holes in anti-de Sitter (AdS) spacetimes surrounded by a cloud of strings (BH-AdS-CoS), incorporating both electric- and magnetic-like components of the string bi-vector. Thermodynamically, these systems exhibit small/intermediate/large black hole phases with first- and second-order transitions governed by the string parameter $c_0$. Dynamically, we probe the phase structure using Lyapunov exponents $λ$ from unstable circular geodesics. For massless particles ($δ= 0$), analytical expressions $λ$ reveal multivalued behavior in first-order transition regimes ($c_0 < c_{\text{cri}}$), with branches mapping to thermodynamic phases ($λ_{\text{SBH}}, λ_{\text{IBH}}, λ_{\text{LBH}}$). The discontinuity $Δλ= λ_{\text{SBH}} - λ_{\text{LBH}}$ at $T_p$ follows mean-field scaling: $Δλ/ λ_{\text{cri}} \propto (T_\text{cri} - T)^{1/2} \quad (β= 1/2)$. For massive particles ($δ= 1$), numerical computation of timelike geodesics confirms $λ$ as an order parameter, with critical exponent $β= 1/2$ universally. Key distinctions emerge: $λ\to 1$ asymptotically for photons, while $λ\to 0$ in the significant black hole phase for massive particles due to vanishing unstable orbits. The transition of $λ$ from multivalued to single-valued at $c_0 = c_{\text{cri}}$ establishes it as a universal dynamical probe of black hole criticality. The universal critical exponent of 1/2 for $Δλ$ further reinforces the analogy with conventional thermodynamic systems. Our results confirm a direct connection between the thermodynamic phase structure of BH-AdS-CoS and the dynamics of test particles, with the Lyapunov exponent emerging as a sensitive diagnostic of black hole criticality.

gr-qc↗

arXiv2025

Recovering Partially Corrupted Objects via Sketch-Guided Bidirectional Feature Interaction

Text-guided diffusion models have achieved remarkable success in object inpainting by providing high-level semantic guidance through text prompts. However, they often lack precise pixel-level spatial control, especially in scenarios involving partially corrupted objects where critical uncorrupted cues remain. To overcome this limitation, sketch-guided methods have been introduced, using either indirect gradient modulation or direct sketch injection to improve structural control. Yet, existing approaches typically establish a one-way mapping from the sketch to the masked regions only, neglecting the contextual information from unmasked object areas. This leads to a disconnection between the sketch and the uncorrupted content, thereby causing sketch-guided inconsistency and structural mismatch. To tackle this challenge, we propose a sketch-guided bidirectional feature interaction framework built upon a pretrained Stable Diffusion model. Our bidirectional interaction features two complementary directions, context-to-sketch and sketch-to-inpainting, that enable fine-grained spatial control for partially corrupted object inpainting. In the context-to-sketch direction, multi-scale latents from uncorrupted object regions are propagated to the sketch branch to generate a visual mask that adapts the sketch features to the visible context and denoising progress. In the sketch-to-inpainting direction, a sketch-conditional affine transformation modulates the influence of sketch guidance based on the learned visual mask, ensuring consistency with uncorrupted object content. This interaction is applied at multiple scales within the encoder of the diffusion U-Net, enabling the model to restore object structures with enhanced spatial fidelity. Extensive experiments on two newly constructed benchmark datasets demonstrate that our approach outperforms state-of-the-art methods.

cs.CV↗

arXiv2025

Epicyclic oscillations and accretion disk around a special Buchdahl-inspired spacetime

In this paper, we consider the Buchdahl-inspired spacetime metric and investigate its aspects using the quasiperiodic oscillations and the accretion disk due to the accreting matter. First, we focus on analyzing the geodesics of particles around the Buchdahl-inspired spacetime, together with the conserved quantities such as specific energy and angular momentum for massive particles orbiting on the innermost stable circular orbits (ISCOs). We show that the effect of the Buchdahl parameter $\tilde{k}$ increases as the radii of the ISCO orbits decrease, resulting in shifting orbits toward the central object compared to the Schwarzschild black hole case. We also consider astrophysical epicyclic oscillations and derive their general expressions using implications of circular motion of massive particles around the Buchdahl-inspired spacetime. Further, we explore the astrophysical implications of observational higher-frequency QPOs of the selected galactic microquasars of X-ray binary systems to obtain the best-fit constraints on the Buchdahl-inspired spacetime parameters. Finally, we consider the accretion disk around the Buchdahl-inspired spacetime using implications of the ISCO parameters that define the accretion disk's inner edge. We explore radiation properties of the accretion disk and redshifted image and intensity of a lensed accretion disk around the Buchdahl-inspired spacetime.

gr-qc↗

arXiv2025

Information Entropy Guided Height-aware Histogram for Quantization-friendly Pillar Feature Encoder

Real-time and high-performance 3D object detection plays a critical role in autonomous driving and robotics. Recent pillar-based 3D object detectors have gained significant attention due to their compact representation and low computational overhead, making them suitable for onboard deployment and quantization. However, existing pillar-based detectors still suffer from information loss along height dimension and large numerical distribution difference during pillar feature encoding (PFE), which severely limits their performance and quantization potential. To address above issue, we first unveil the importance of different input information during PFE and identify the height dimension as a key factor in enhancing 3D detection performance. Motivated by this observation, we propose a height-aware pillar feature encoder, called PillarHist. Specifically, PillarHist statistics the discrete distribution of points at different heights within one pillar with the information entropy guidance. This simple yet effective design greatly preserves the information along the height dimension while significantly reducing the computation overhead of the PFE. Meanwhile, PillarHist also constrains the arithmetic distribution of PFE input to a stable range, making it quantization-friendly. Notably, PillarHist operates exclusively within the PFE stage to enhance performance, enabling seamless integration into existing pillar-based methods without introducing complex operations. Extensive experiments show the effectiveness of PillarHist in terms of both efficiency and performance.

cs.CV↗

arXiv2025

JointSplat: Probabilistic Joint Flow-Depth Optimization for Sparse-View Gaussian Splatting

Reconstructing 3D scenes from sparse viewpoints is a long-standing challenge with wide applications. Recent advances in feed-forward 3D Gaussian sparse-view reconstruction methods provide an efficient solution for real-time novel view synthesis by leveraging geometric priors learned from large-scale multi-view datasets and computing 3D Gaussian centers via back-projection. Despite offering strong geometric cues, both feed-forward multi-view depth estimation and flow-depth joint estimation face key limitations: the former suffers from mislocation and artifact issues in low-texture or repetitive regions, while the latter is prone to local noise and global inconsistency due to unreliable matches when ground-truth flow supervision is unavailable. To overcome this, we propose JointSplat, a unified framework that leverages the complementarity between optical flow and depth via a novel probabilistic optimization mechanism. Specifically, this pixel-level mechanism scales the information fusion between depth and flow based on the matching probability of optical flow during training. Building upon the above mechanism, we further propose a novel multi-view depth-consistency loss to leverage the reliability of supervision while suppressing misleading gradients in uncertain areas. Evaluated on RealEstate10K and ACID, JointSplat consistently outperforms state-of-the-art (SOTA) methods, demonstrating the effectiveness and robustness of our proposed probabilistic joint flow-depth optimization approach for high-fidelity sparse-view 3D reconstruction.

cs.CV↗

arXiv2025

PersonaX: A Recommendation Agent Oriented User Modeling Framework for Long Behavior Sequence

User profile embedded in the prompt template of personalized recommendation agents play a crucial role in shaping their decision-making process. High-quality user profiles are essential for aligning agent behavior with real user interests. Typically, these profiles are constructed by leveraging LLMs for user profile modeling (LLM-UM). However, this process faces several challenges: (1) LLMs struggle with long user behaviors due to context length limitations and performance degradation. (2) Existing methods often extract only partial segments from full historical behavior sequence, inevitably discarding diverse user interests embedded in the omitted content, leading to incomplete modeling and suboptimal profiling. (3) User profiling is often tightly coupled with the inference context, requiring online processing, which introduces significant latency overhead. In this paper, we propose PersonaX, an agent-agnostic LLM-UM framework to address these challenges. It augments downstream recommendation agents to achieve better recommendation performance and inference efficiency. PersonaX (a) segments complete historical behaviors into clustered groups, (b) selects multiple sub behavior sequences (SBS) with a balance of prototypicality and diversity to form a high quality core set, (c) performs offline multi-persona profiling to capture diverse user interests and generate fine grained, cached textual personas, and (d) decouples user profiling from online inference, enabling profile retrieval instead of real time generation. Extensive experiments demonstrate its effectiveness: using only 30 to 50% of behavioral data (sequence length 480), PersonaX enhances AgentCF by 3 to 11% and Agent4Rec by 10 to 50%. As a scalable and model-agnostic LLM-UM solution, PersonaX sets a new benchmark in scalable user modeling.

cs.IR↗

arXiv2025

FedPCL-CDR: A Federated Prototype-based Contrastive Learning Framework for Privacy-Preserving Cross-domain Recommendation

Cross-domain recommendation (CDR) aims to improve recommendation accuracy in sparse domains by transferring knowledge from data-rich domains. However, existing CDR approaches often assume that user-item interaction data across domains is publicly available, neglecting user privacy concerns. Additionally, they experience performance degradation with sparse overlapping users due to their reliance on a large number of fully shared users for knowledge transfer. To address these challenges, we propose a Federated Prototype-based Contrastive Learning (CL) framework for Privacy Preserving CDR, called FedPCL-CDR. This approach utilizes non-overlapping user information and differential prototypes to improve model performance within a federated learning framework. FedPCL-CDR comprises two key modules: local domain (client) learning and global server aggregation. In the local domain, FedPCL-CDR first clusters all user data and utilizes local differential privacy (LDP) to learn differential prototypes, effectively utilizing non-overlapping user information and protecting user privacy. It then conducts knowledge transfer by employing both local and global prototypes returned from the server in a CL manner. Meanwhile, the global server aggregates differential prototypes sent from local domains to learn both local and global prototypes. Extensive experiments on four CDR tasks across Amazon and Douban datasets demonstrate that FedPCL-CDR surpasses SOTA baselines. We release our code at https://github.com/Lili1013/FedPCL CDR

cs.IR↗

arXiv2025

Starobinsky like inflation and EGB Gravity in the light of ACT

The ACT data shows an enhancement in the value of the scalar spectral index, as $n_s = 0.9743 \pm 0.0034$, leading to disfavoring many inflationary models, including the Starobinsky model. To satisfy the constraint made by ACT, we will investigate the Starobinsky potential within the Einstein-Gauss-Bonnet(EGB) gravity theory. EGB gravity is motivated by the higher-dimensional theory, which includes quadratic curvature correction terms and a coupling between the scalar field and the GB term that modifies the dynamical equations. The model is considered by using the slow-roll approximation method and the exact numerical approach for two different coupling functions. The results indicate that the model is in good agreement with the data and the results stand in $1σ$ of the ACT $r-n_s$ plane. Considering the running of the scalar spectral index also implies the consistency of the model with data. In addition, the parametric space of the free parameters of the EGB coupling is explored, where we find the acceptable region of the parameters in which the resulting $n_s$ and $r$ stay in $1σ$ of the ACT data. Next, the reheating phase is considered. It is determined that the model can simultaneously satisfy the constraint of ACT data and the reheating temperature constraints.

astro-ph.CO↗

arXiv2025

Enhancing News Recommendation with Hierarchical LLM Prompting

Personalized news recommendation systems often struggle to effectively capture the complexity of user preferences, as they rely heavily on shallow representations, such as article titles and abstracts. To address this problem, we introduce a novel method, namely PNR-LLM, for Large Language Models for Personalized News Recommendation. Specifically, PNR-LLM harnesses the generation capabilities of LLMs to enrich news titles and abstracts, and consequently improves recommendation quality. PNR-LLM contains a novel module, News Enrichment via LLMs, which generates deeper semantic information and relevant entities from articles, transforming shallow contents into richer representations. We further propose an attention mechanism to aggregate enriched semantic- and entity-level data, forming unified user and news embeddings that reveal a more accurate user-news match. Extensive experiments on MIND datasets show that PNR-LLM outperforms state-of-the-art baselines. Moreover, the proposed data enrichment module is model-agnostic, and we empirically show that applying our proposed module to multiple existing models can further improve their performance, verifying the advantage of our design.

cs.IR↗

arXiv2025

The effects of asymptotically flat $R^2$ spacetime on black hole image of Sagittarius A*

A new class of analytically expressible vacuum solutions has recently been discovered for pure ${R}^2$ gravity, building upon Buchdahl's seminal work from 1962. These solutions, inspired by Buchdahl's framework, offer a promising avenue for testing ${R}^2$ gravity against astrophysical observations. Within a subset of asymptotically flat Buchdahl-inspired vacuum spacetimes, we introduce a free parameter $ε$ to characterize deviations from the Schwarzschild metric, which is recovered in the limit $ε= 0$. In this study, we employ the publicly available code \textit{ipole} to simulate black hole images under the Buchdahl-inspired metric, with a focus on the black hole at the center of the Milky Way, Sagittarius A* (Sgr A*). Our simulations show that both the shadow size and photon ring diameter decrease monotonically with increasing $ε$. By exploring a range of observational inclination angles, we find that the photon ring diameter being a direct observable is only weakly sensitive to the inclination angle. We further constrain the parameter $ε$ by comparing our simulation results with the Event Horizon Telescope (EHT) observations of Sgr A*. The obtained bounds are consistent with those previously derived from the orbital motion of the S2 star, but provide tighter constraints. In addition, we analyze the influence of the Buchdahl-inspired spacetime on the polarization patterns near the black hole and find its impact to be minimal. In contrast, the observational inclination angle has a substantial effect on the observed polarization structure, highlighting the dominant role of viewing geometry in shaping polarization features.

gr-qc↗

arXiv2025

ANNs-SaDE: A Machine-Learning-Based Design Automation Framework for Microwave Branch-Line Couplers

The traditional method for designing branch-line couplers involves a trial-and-error optimization process that requires multiple design iterations through electromagnetic (EM) simulations. Thus, it is extremely time consuming and labor intensive. In this paper, a novel machine-learning-based framework is proposed to tackle this issue. It integrates artificial neural networks with a self-adaptive differential evolution algorithm (ANNs-SaDE). This framework enables the self-adaptive design of various types of microwave branch-line couplers by precisely optimizing essential electrical properties, such as coupling factor, isolation, and phase difference between output ports. The effectiveness of the ANNs-SaDE framework is demonstrated by the designs of folded single-stage branch-line couplers and multi-stage wideband branch-line couplers.

eess.SP↗

arXiv2025

A Parallel Hybrid Action Space Reinforcement Learning Model for Real-world Adaptive Traffic Signal Control

Adaptive traffic signal control (ATSC) can effectively reduce vehicle travel times by dynamically adjusting signal timings but poses a critical challenge in real-world scenarios due to the complexity of real-time decision-making in dynamic and uncertain traffic conditions. The burgeoning field of intelligent transportation systems, bolstered by artificial intelligence techniques and extensive data availability, offers new prospects for the implementation of ATSC. In this study, we introduce a parallel hybrid action space reinforcement learning model (PH-DDPG) that optimizes traffic signal phase and duration of traffic signals simultaneously, eliminating the need for sequential decision-making seen in traditional two-stage models. Our model features a task-specific parallel hybrid action space tailored for adaptive traffic control, which directly outputs discrete phase selections and their associated continuous duration parameters concurrently, thereby inherently addressing dynamic traffic adaptation through unified parametric optimization. %Our model features a unique parallel hybrid action space that allows for the simultaneous output of each action and its optimal parameters, streamlining the decision-making process. Furthermore, to ascertain the robustness and effectiveness of this approach, we executed ablation studies focusing on the utilization of a random action parameter mask within the critic network, which decouples the parameter space for individual actions, facilitating the use of preferable parameters for each action. The results from these studies confirm the efficacy of this method, distinctly enhancing real-world applicability

cs.AI↗

arXiv2025

Sampling from the Continuous Random Energy Model in Total Variation Distance

The continuous random energy model (CREM) is a toy model of spin glasses on $\{0,1\}^N$ that, in the limit, exhibits an infinitely hierarchical correlation structure. We give two polynomial-time algorithms to approximately sample from the Gibbs distribution of the CREM in the high-temperature regime $β<β_{\min}:=\min\{β_c,β_G\}$, based on a Markov chain and a sequential sampler. The running time depends algebraically on the desired TV distance and failure probability and exponentially in $(1/g)^{O(1)}$, where $g$ is the gap to a certain inverse temperature threshold $β_{\min}$; this contrasts with previous results which only attain $o(N)$ accuracy in KL divergence. If the covariance function $A$ of the CREM is concave, the algorithms work up to the critical threshold $β_c$, which is the static phase transition point; while for $A$ non-concave, if $β_G<β_c$, the algorithms work up to the known algorithmic threshold $β_G$ proposed in Addario-Berry and Maillard (2020) for non-trivial sampling guarantees. Our result depends on quantitative bounds for the fluctuation of the partition function and a new contiguity result of the ``tilted" CREM obtained from sampling, which is of independent interest. We also show that the spectral gap is exponentially small with high probability, suggesting that the algebraic dependence is unavoidable with a Markov chain approach.

math.PR↗

arXiv2025

Merino: Entropy-driven Design for Generative Language Models on IoT Devices

Generative Large Language Models (LLMs) stand as a revolutionary advancement in the modern era of artificial intelligence (AI). However, scaling down LLMs for resource-constrained hardware, such as Internet-of-Things (IoT) devices requires non-trivial efforts and domain knowledge. In this paper, we propose a novel information-entropy framework for designing mobile-friendly generative language models. The whole design procedure involves solving a mathematical programming (MP) problem, which can be done on the CPU within minutes, making it nearly zero-cost. We evaluate our designed models, termed MeRino, across fourteen NLP downstream tasks, showing their competitive performance against the state-of-the-art autoregressive transformer models under the mobile setting. Notably, MeRino achieves similar or better performance on both language modeling and zero-shot learning tasks, compared to the 350M parameter OPT while being 4.9x faster on NVIDIA Jetson Nano with 5.5x reduction in model size.

cs.LG↗

arXiv2024

TSUBF-Net: Trans-Spatial UNet-like Network with Bi-direction Fusion for Segmentation of Adenoid Hypertrophy in CT

Adenoid hypertrophy stands as a common cause of obstructive sleep apnea-hypopnea syndrome in children. It is characterized by snoring, nasal congestion, and growth disorders. Computed Tomography (CT) emerges as a pivotal medical imaging modality, utilizing X-rays and advanced computational techniques to generate detailed cross-sectional images. Within the realm of pediatric airway assessments, CT imaging provides an insightful perspective on the shape and volume of enlarged adenoids. Despite the advances of deep learning methods for medical imaging analysis, there remains an emptiness in the segmentation of adenoid hypertrophy in CT scans. To address this research gap, we introduce TSUBF-Nett (Trans-Spatial UNet-like Network based on Bi-direction Fusion), a 3D medical image segmentation framework. TSUBF-Net is engineered to effectively discern intricate 3D spatial interlayer features in CT scans and enhance the extraction of boundary-blurring features. Notably, we propose two innovative modules within the U-shaped network architecture:the Trans-Spatial Perception module (TSP) and the Bi-directional Sampling Collaborated Fusion module (BSCF).These two modules are in charge of operating during the sampling process and strategically fusing down-sampled and up-sampled features, respectively. Furthermore, we introduce the Sobel loss term, which optimizes the smoothness of the segmentation results and enhances model accuracy. Extensive 3D segmentation experiments are conducted on several datasets. TSUBF-Net is superior to the state-of-the-art methods with the lowest HD95: 7.03, IoU:85.63, and DSC: 92.26 on our own AHSD dataset. The results in the other two public datasets also demonstrate that our methods can robustly and effectively address the challenges of 3D segmentation in CT scans.

eess.IV↗

arXiv2024

Peri-midFormer: Periodic Pyramid Transformer for Time Series Analysis

Time series analysis finds wide applications in fields such as weather forecasting, anomaly detection, and behavior recognition. Previous methods attempted to model temporal variations directly using 1D time series. However, this has been quite challenging due to the discrete nature of data points in time series and the complexity of periodic variation. In terms of periodicity, taking weather and traffic data as an example, there are multi-periodic variations such as yearly, monthly, weekly, and daily, etc. In order to break through the limitations of the previous methods, we decouple the implied complex periodic variations into inclusion and overlap relationships among different level periodic components based on the observation of the multi-periodicity therein and its inclusion relationships. This explicitly represents the naturally occurring pyramid-like properties in time series, where the top level is the original time series and lower levels consist of periodic components with gradually shorter periods, which we call the periodic pyramid. To further extract complex temporal variations, we introduce self-attention mechanism into the periodic pyramid, capturing complex periodic relationships by computing attention between periodic components based on their inclusion, overlap, and adjacency relationships. Our proposed Peri-midFormer demonstrates outstanding performance in five mainstream time series analysis tasks, including short- and long-term forecasting, imputation, classification, and anomaly detection.

cs.LG↗

arXiv2024

Soft-Matter-Based Topological Vertical Cavity Surface Emitting Lasers

Polarized topological vertical cavity surface-emitting lasers (VCSELs), as stable and efficient on-chip light sources, play an important role in the next generation of optical storage and optical communications. However, most current topological lasers demand complex design and expensive fabrication processes, and their semiconductor-based structures pose challenges for flexible device applications. By use of an analogy with two-dimensional Semenov insulators in synthetic parametric space, we design and realize a one-dimensional optical superlattice (stacked polymerized cholesteric liquid crystal films and Mylar films), thereby we demonstrate a flexible, low threshold, circularly polarized topological VCSEL with high slope efficiency. We show that such a laser maintains a good single-mode property under low pump power and inherits the transverse spatial profile of the pump laser. Thanks to the soft-matter-based flexibility, our topological VCSEL can be "attached" to substrates of various shapes, enabling desired laser properties and robust beam steering even after undergoing hundreds of bends. Our results may find applications in consumer electronics, laser scanning and displays, as well as wearable devices.

physics.optics↗