arXiv subjects

Di Wu

Publications and source records attributed to Di Wu.

At least 307 records · Page 17Linked to original sources

arXiv2025

EMO: Edge Model Overlays to Scale Model Size in Federated Learning

Federated Learning (FL) trains machine learning models on edge devices with distributed data. However, the computational and memory limitations of these devices restrict the training of large models using FL. Split Federated Learning (SFL) addresses this challenge by distributing the model across the device and server, but it introduces a tightly coupled data flow, leading to computational bottlenecks and high communication costs. We propose EMO as a solution to enable the training of large models in FL while mitigating the challenges of SFL. EMO introduces Edge Model Overlay(s) between the device and server, enabling the creation of a larger ensemble model without modifying the FL workflow. The key innovation in EMO is Augmented Federated Learning (AFL), which builds an ensemble model by connecting the original (smaller) FL model with model(s) trained in the overlay(s) to facilitate horizontal or vertical scaling. This is accomplished through three key modules: a hierarchical activation replay cache to decouple AFL from FL, a convergence-aware communication controller to optimize communication overhead, and an ensemble inference module. Evaluations on a real-world prototype show that EMO improves accuracy by up to 17.77% compared to FL, and reduces communication costs by up to 7.17x and decreases training time by up to 6.9x compared to SFL.

cs.LG↗

arXiv2025

Can tensor-scalar induced GWs dominate PTA observations ?

Observational constraints on small-scale primordial gravitational waves are considerably weaker than those on large scales. We focus on scenarios with significant primordial gravitational waves and curvature perturbations on small scales, studying the energy density spectrum of the second-order TSIGW. By leveraging current data from CMB, BAO, and PTA, combined with the SNR analysis of LISA, we can investigate how tensor-scalar induced gravitational waves affect observations on various scales, thus constraining the parameter space for primordial gravitational waves and curvature perturbations. The Bayes factor analysis suggests that TSIGW+PGW might be more likely to dominate current PTA observations compared to SMBHB.

astro-ph.CO↗

arXiv2025

Analytical Strategies and Winning Conditions for Elliptic-Orbit Target-Attacker-Defender Game

This paper proposes an analytical framework for the orbital Target-Attacker-Defender game with a non-maneuvering target along elliptic orbits. Focusing on the linear quadratic game, we derive an analytical solution to the matrix Riccati equation, which yields analytical Nash-equilibrium strategies for the game. Based on the analytical strategies, we derive the analytical form of the necessary and sufficient winning conditions for the attacker. The simulation results show good consistency between the analytical and numerical methods, exhibiting 0.004$\%$ relative error in the cost function. The analytical method achieves over 99.9$\%$ reduction in CPU time compared to the conventional numerical method, strengthening the advantage of developing the analytical strategies. Furthermore, we verify the proposed winning conditions and investigate the effects of eccentricity on the game outcomes. Our analysis reveals that for games with hovering initial states, the initial position of the defender should be constrained inside a mathematically definable set to ensure that the attacker wins the game. This constrained set further permits geometric interpretation through our proposed method. This work establishes the analytical framework for orbital Target-Attacker-Defender games, providing fundamental insights into the solution analysis of the game.

math.OC↗

arXiv2025

Accelerate High-Quality Diffusion Models with Inner Loop Feedback

We propose Inner Loop Feedback (ILF), a novel approach to accelerate diffusion models' inference. ILF trains a lightweight module to predict future features in the denoising process by leveraging the outputs from a chosen diffusion backbone block at a given time step. This approach exploits two key intuitions; (1) the outputs of a given block at adjacent time steps are similar, and (2) performing partial computations for a step imposes a lower burden on the model than skipping the step entirely. Our method is highly flexible, since we find that the feedback module itself can simply be a block from the diffusion backbone, with all settings copied. Its influence on the diffusion forward can be tempered with a learnable scaling factor from zero initialization. We train this module using distillation losses; however, unlike some prior work where a full diffusion backbone serves as the student, our model freezes the backbone, training only the feedback module. While many efforts to optimize diffusion models focus on achieving acceptable image quality in extremely few steps (1-4 steps), our emphasis is on matching best case results (typically achieved in 20 steps) while significantly reducing runtime. ILF achieves this balance effectively, demonstrating strong performance for both class-to-image generation with diffusion transformer (DiT) and text-to-image generation with DiT-based PixArt-alpha and PixArt-sigma. The quality of ILF's 1.7x-1.8x speedups are confirmed by FID, CLIP score, CLIP Image Quality Assessment, ImageReward, and qualitative comparisons. Project information is available at https://mgwillia.github.io/ilf.

cs.CV↗

arXiv2025

Induced gravitational waves for arbitrary higher orders: vertex rules and loop diagrams in cosmological perturbation theory

Gravitational waves induced by primordial perturbations serve as crucial probes for studying the early universe, providing a significant window into potential new physics during cosmic evolution. Due to the potentially large amplitudes of primordial perturbations on small scales, the contributions of high-order cosmological perturbations are highly significant. We propose a vertex approach applicable to the study of induced gravitational waves for arbitrary higher orders. Using the vertex approach and tree diagrams, we can directly derive the explicit expressions of higher-order induced gravitational waves without involving the complex and lengthy calculations of higher-order cosmological perturbations. Correlations between different tree diagrams correspond to the loop diagrams of two-point correlation functions of induced gravitational waves. Our investigation reveals that one-particle reducible diagrams impact tensor-scalar induced gravitational waves while leaving scalar induced gravitational waves unaffected.

astro-ph.CO↗

arXiv2025

Generative AI as a Service in 6G Edge-Cloud: Generation Task Offloading by In-context Learning

Generative artificial intelligence (GAI) is a promising technique towards 6G networks, and generative foundation models such as large language models (LLMs) have attracted considerable interest from academia and telecom industry. This work considers a novel edge-cloud deployment of foundation models in 6G networks. Specifically, it aims to minimize the service delay of foundation models by radio resource allocation and task offloading, i.e., offloading diverse content generation tasks to proper LLMs at the network edge or cloud. In particular, we first introduce the communication system model, i.e., allocating radio resources and calculating link capacity to support generated content transmission, and then we present the LLM inference model to calculate the delay of content generation. After that, we propose a novel in-context learning method to optimize the task offloading decisions. It utilizes LLM's inference capabilities, and avoids the difficulty of dedicated model training or fine-tuning as in conventional machine learning algorithms. Finally, the simulations demonstrate that the proposed edge-cloud deployment and in-context learning task offloading method can achieve satisfactory generation service quality without dedicated model training or fine-tuning.

eess.SY↗

arXiv2025

Giant energy density nitride dielectrics enabled by a paraelectric-metaparaelectric phase transition

Electrostatic dielectric capacitors are foundational to advance the electronics and electric power devices due to their ultrafast charging/discharging capability and high-power density. However, the low energy density limits the potential for next generation devices in terms of miniaturization and integration. We propose a strategy that relies on inducing a field-driven phase transition that we denote paraelectric-metaparaelectric, which yields an ultrahigh energy density in III-nitrides. III-nitride compounds (Al, Sc, B)N with certain cation concentrations possess a nonpolar hexagonal ground phase which could transform into a polar wurtzite phase under a very large electric field, which is denoted as metaparaelectric with nearly null hysteresis P-E loop. This paraelectric-metaparaelectric transition leads to a polarization saturation at large electric field. The corresponding P-E loop displays a giant energy density of 308 J/cm$^3$ with high efficiency nearly 100%. The proposed paraelectric-metaparaelectric phase transition strategy in nitrides opens an avenue to design of next generation high performance dielectrics.

cond-mat.mtrl-sci↗

arXiv2025

Sampling Innovation-Based Adaptive Compressive Sensing

Scene-aware Adaptive Compressive Sensing (ACS) has attracted significant interest due to its promising capability for efficient and high-fidelity acquisition of scene images. ACS typically prescribes adaptive sampling allocation (ASA) based on previous samples in the absence of ground truth. However, when confronting unknown scenes, existing ACS methods often lack accurate judgment and robust feedback mechanisms for ASA, thus limiting the high-fidelity sensing of the scene. In this paper, we introduce a Sampling Innovation-Based ACS (SIB-ACS) method that can effectively identify and allocate sampling to challenging image reconstruction areas, culminating in high-fidelity image reconstruction. An innovation criterion is proposed to judge ASA by predicting the decrease in image reconstruction error attributable to sampling increments, thereby directing more samples towards regions where the reconstruction error diminishes significantly. A sampling innovation-guided multi-stage adaptive sampling (AS) framework is proposed, which iteratively refines the ASA through a multi-stage feedback process. For image reconstruction, we propose a Principal Component Compressed Domain Network (PCCD-Net), which efficiently and faithfully reconstructs images under AS scenarios. Extensive experiments demonstrate that the proposed SIB-ACS method significantly outperforms the state-of-the-art methods in terms of image reconstruction fidelity and visual effects. Codes are available at https://github.com/giant-pandada/SIB-ACS_CVPR2025.

cs.CV↗

arXiv2025

FedGAI: Federated Style Learning with Cloud-Edge Collaboration for Generative AI in Fashion Design

Collaboration can amalgamate diverse ideas, styles, and visual elements, fostering creativity and innovation among different designers. In collaborative design, sketches play a pivotal role as a means of expressing design creativity. However, designers often tend to not openly share these meticulously crafted sketches. This phenomenon of data island in the design area hinders its digital transformation under the third wave of AI. In this paper, we introduce a Federated Generative Artificial Intelligence Clothing system, namely FedGAI, employing federated learning to aid in sketch design. FedGAI is committed to establishing an ecosystem wherein designers can exchange sketch styles among themselves. Through FedGAI, designers can generate sketches that incorporate various designers' styles from their peers, drawing inspiration from collaboration without the need for data disclosure or upload. Extensive performance evaluations indicate that our FedGAI system can produce multi-styled sketches of comparable quality to human-designed ones while significantly enhancing efficiency compared to hand-drawn sketches.

cs.AI↗

arXiv2025

Cost-Aware Optimal Pairwise Pure Exploration

Pure exploration is one of the fundamental problems in multi-armed bandits (MAB). However, existing works mostly focus on specific pure exploration tasks, without a holistic view of the general pure exploration problem. This work fills this gap by introducing a versatile framework to study pure exploration, with a focus on identifying the pairwise relationships between targeted arm pairs. Moreover, unlike existing works that only optimize the stopping time (i.e., sample complexity), this work considers that arms are associated with potentially different costs and targets at optimizing the cumulative cost that occurred during learning. Under the general framework of pairwise pure exploration with arm-specific costs, a performance lower bound is derived. Then, a novel algorithm, termed CAET (Cost-Aware Pairwise Exploration Task), is proposed. CAET builds on the track-and-stop principle with a novel design to handle the arm-specific costs, which can potentially be zero and thus represent a very challenging case. Theoretical analyses prove that the performance of CAET approaches the lower bound asymptotically. Special cases are further discussed, including an extension to regret minimization, which is another major focus of MAB. The effectiveness and efficiency of CAET are also verified through experimental results under various settings.

stat.ML↗

arXiv2025

LongMemEval: Benchmarking Chat Assistants on Long-Term Interactive Memory

Recent large language model (LLM)-driven chat assistant systems have integrated memory components to track user-assistant chat histories, enabling more accurate and personalized responses. However, their long-term memory capabilities in sustained interactions remain underexplored. We introduce LongMemEval, a comprehensive benchmark designed to evaluate five core long-term memory abilities of chat assistants: information extraction, multi-session reasoning, temporal reasoning, knowledge updates, and abstention. With 500 meticulously curated questions embedded within freely scalable user-assistant chat histories, LongMemEval presents a significant challenge to existing long-term memory systems, with commercial chat assistants and long-context LLMs showing a 30% accuracy drop on memorizing information across sustained interactions. We then present a unified framework that breaks down the long-term memory design into three stages: indexing, retrieval, and reading. Built upon key experimental insights, we propose several memory design optimizations including session decomposition for value granularity, fact-augmented key expansion for indexing, and time-aware query expansion for refining the search scope. Extensive experiments show that these optimizations greatly improve both memory recall and downstream question answering on LongMemEval. Overall, our study provides valuable resources and guidance for advancing the long-term memory capabilities of LLM-based chat assistants, paving the way toward more personalized and reliable conversational AI. Our benchmark and code are publicly available at https://github.com/xiaowu0162/LongMemEval.

cs.CL↗

arXiv2025

Optimal Power Management for Large-Scale Battery Energy Storage Systems via Bayesian Inference

Large-scale battery energy storage systems (BESS) have found ever-increasing use across industry and society to accelerate clean energy transition and improve energy supply reliability and resilience. However, their optimal power management poses significant challenges: the underlying high-dimensional nonlinear nonconvex optimization lacks computational tractability in real-world implementation, and the uncertainty of the exogenous power demand makes exact optimization difficult. This paper presents a new solution framework to address these bottlenecks. The solution pivots on introducing power-sharing ratios to specify each cell's power quota from the output power demand. To find the optimal power-sharing ratios, we formulate a nonlinear model predictive control (NMPC) problem to achieve power-loss-minimizing BESS operation while complying with safety, cell balancing, and power supply-demand constraints. We then propose a parameterized control policy for the power-sharing ratios, which utilizes only three parameters, to reduce the computational demand in solving the NMPC problem. This policy parameterization allows us to translate the NMPC problem into a Bayesian inference problem for the sake of 1) computational tractability, and 2) overcoming the nonconvexity of the optimization problem. We leverage the ensemble Kalman inversion technique to solve the parameter estimation problem. Concurrently, a low-level control loop is developed to seamlessly integrate our proposed approach with the BESS to ensure practical implementation. This low-level controller receives the optimal power-sharing ratios, generates output power references for the cells, and maintains a balance between power supply and demand despite uncertainty in output power. We conduct extensive simulations and experiments on a 20-cell prototype to validate the proposed approach.

eess.SY↗

arXiv2025

The Kaluza-Klein AdS Virasoro-Shapiro Amplitude near Flat Space

We bootstrap the first-order correction in the curvature expansion of the Virasoro-Shapiro amplitude in AdS spacetime, for arbitrary Kaluza-Klein charges of external operators. By constructing a universal ansatz based on single-valued multiple polylogarithms as well as an AdS$\times$S formalism, and matching it with the low-lying result, we derive a unified formula in terms of world-sheet integrals. Our result predicts an infinite number of Wilson coefficients that were not available in previous literature.

hep-th↗

arXiv2025

Efficient Fault Diagnosis in Lithium-Ion Battery Packs: A Structural Approach with Moving Horizon Estimation

Safe and reliable operation of lithium-ion battery packs depends on effective fault diagnosis. However, model-based approaches often encounter two major challenges: high computational complexity and extensive sensor requirements. To address these bottlenecks, this paper introduces a novel approach that harnesses the structural properties of battery packs, including cell uniformity and the sparsity of fault occurrences. We integrate this approach into a Moving Horizon Estimation (MHE) framework and estimate fault signals such as internal and external short circuits and faults in voltage and current sensors. To mitigate computational demands, we propose a hierarchical solution to the MHE problem. The proposed solution breaks up the pack-level MHE problem into smaller problems and solves them efficiently. Finally, we perform extensive simulations across various battery pack configurations and fault types to demonstrate the effectiveness of the proposed approach. The results highlight that the proposed approach simultaneously reduces the computational demands and sensor requirements of fault diagnosis.

eess.SY↗

arXiv2025

Novel topological classes in black hole thermodynamics

By viewing black hole solutions as topological defects in thermodynamic parameter space, we unveil a novel topological class and two new topological subclasses, respectively denoted $W^{0-\leftrightarrow 1+}$, $\overline{W}^{1+}$, and $\hat{W}^{1+}$, that extend beyond the four established categories proposed by Wei et al. [Phys. Rev. D 110, L081501 (2024)]. Within the newly identified class and these two novel subclasses, the innermost small black hole states exhibit a distinct sequence of unstable, stable, and stable behaviours, while the outermost large black hole states display a uniform pattern of stable behaviours. These classifications indicate thermodynamic properties in both the low and high Hawking temperature regimes that are strikingly different from the previously known four topological classes. In particular, we demonstrate that static charged AdS black holes in gauged supergravity exhibit an intricate thermodynamic evolution that is notably distinct from that of the Reissner-Nordström anti-de Sitter (RN-AdS) black hole. From a topological perspective, we emphasize the advantages and potential of investigating thermodynamic phase transitions in these black hole spacetimes, an area that has been rarely explored in previous research. Our findings not only enrich and sharpen the framework of topological classifications in black hole thermodynamics, but also represent a significant stride toward unraveling the fundamental nature of black holes and gravity.

hep-th↗

arXiv2025

Towards Homogeneous Lexical Tone Decoding from Heterogeneous Intracranial Recordings

Recent advancements in brain-computer interfaces (BCIs) have enabled the decoding of lexical tones from intracranial recordings, offering the potential to restore the communication abilities of speech-impaired tonal language speakers. However, data heterogeneity induced by both physiological and instrumental factors poses a significant challenge for unified invasive brain tone decoding. Traditional subject-specific models, which operate under a heterogeneous decoding paradigm, fail to capture generalized neural representations and cannot effectively leverage data across subjects. To address these limitations, we introduce Homogeneity-Heterogeneity Disentangled Learning for neural Representations (H2DiLR), a novel framework that disentangles and learns both the homogeneity and heterogeneity from intracranial recordings across multiple subjects. To evaluate H2DiLR, we collected stereoelectroencephalography (sEEG) data from multiple participants reading Mandarin materials comprising 407 syllables, representing nearly all Mandarin characters. Extensive experiments demonstrate that H2DiLR, as a unified decoding paradigm, significantly outperforms the conventional heterogeneous decoding approach. Furthermore, we empirically confirm that H2DiLR effectively captures both homogeneity and heterogeneity during neural representation learning.

cs.CL↗

arXiv2025

The Sliding Flux Ramp Demodulation Algorithm with High Sampling Rate in Microwave SQUID Multiplexer

Microwave SQUID Multiplexing (uMUX) is a widely used technique in the low temperature detectors community as it offers high capacity of reading out large scale Transition-Edge Sensor (TES) arrays. In this paper, we propose a Sliding Flux Ramp Demodulation (SFRD) algorithm for uMUX readout system. It can achieve a sampling rate in the order of MHz while maintaining a multiplexing ratio about one thousand. Advancing of this large array readout technique makes it possible to observe scientiffc objects with improved time resolution and event count rate. This will be highly helpful for TES calorimeters in X-ray applications, such as X-ray astrophysics missions.

hep-ex↗

arXiv2025

Large Memory Network for Recommendation

Modeling user behavior sequences in recommender systems is essential for understanding user preferences over time, enabling personalized and accurate recommendations for improving user retention and enhancing business values. Despite its significance, there are two challenges for current sequential modeling approaches. From the spatial dimension, it is difficult to mutually perceive similar users' interests for a generalized intention understanding; from the temporal dimension, current methods are generally prone to forgetting long-term interests due to the fixed-length input sequence. In this paper, we present Large Memory Network (LMN), providing a novel idea by compressing and storing user history behavior information in a large-scale memory block. With the elaborated online deployment strategy, the memory block can be easily scaled up to million-scale in the industry. Extensive offline comparison experiments, memory scaling up experiments, and online A/B test on Douyin E-Commerce Search (ECS) are performed, validating the superior performance of LMN. Currently, LMN has been fully deployed in Douyin ECS, serving millions of users each day.

cs.IR↗