arXiv subjects

Weifeng Zhang

Publications and source records attributed to Weifeng Zhang.

At least 19 recordsLinked to original sources

arXiv2026

DisasterBench: Benchmarking LLM Planning under Typed Tool Interface Constraints

Disasters cause severe societal impacts, demanding rapid coordination of heterogeneous AI tools, from satellite analysis to flood prediction and damage assessment, into coherent multi-step workflows. As LLMs increasingly serve as orchestrators of such pipelines, effective coordination requires more than selecting semantically plausible tools: LLMs must generate executable workflows with correct parameter binding and dependency propagation. We introduce DisasterBench, a benchmark for evaluating structured multi-agent planning over semantically similar but operationally distinct disaster-response tools. To enable step-level failure attribution, we further propose First-Point-of-Failure (FPoF), which localizes the earliest root cause in a predicted workflow, separating primary errors from downstream cascading effects. Our evaluation reveals three findings: planning method effectiveness depends strongly on model capacity; tool mismatch and parameter-binding errors dominate first failures, revealing semantic grounding and execution consistency as distinct bottlenecks; and verbose intermediate reasoning can create instruction clash with structured output requirements, disrupting plan generation. Together, these findings highlight a fundamental gap between semantic reasoning and execution-grounded coordination, underscoring the need for planning frameworks that jointly model semantic intent, execution constraints, and workflow consistency. Code, data, and evaluation resources are available at: https://github.com/TamuChen18/DisasterBench_Open

cs.CL↗

arXiv2026

NasZip: Software and Hardware Co-Design to Accelerate Approximate Nearest Neighbor Search with DIMM-Based Near-Data Processing

As large language models (LLMs) continue to advance, retrieval-augmented generation (RAG) has become the key mechanism for expanding model knowledge and reducing hallucinations. Central to RAG is approximate nearest neighbor search (ANNS), which retrieves database vectors most similar to a given query. However, distance calculation over high-dimensional vectors is inherently memory-bound, causing retrieval performance to be constrained by I/O bandwidth on mainstream platforms such as CPUs and GPUs. Although many prior early exiting (EE) techniques attempt to reduce memory accesses by only computing partial dimensions, the partial distance converges too slowly to the EE threshold, which ultimately limits their performance gains. To address these challenges, we propose NASZIP, a hardware-software co-designed framework that integrates near data processing (NDP) with a novel feature-level early exiting guided by statistics-based principal component analysis (PCA). Instead of relying solely on partial distances, NASZIP incorporates estimation and correction parameters to approximate full dimensional distances accurately, enabling earlier exiting without compromising accuracy. We further introduce a bit-level NDP-aware dynamic-float scheme that significantly reduces memory access for vector data. On the hardware side, we develop a data aware neighbor list mapping strategy that reduces neighbor retrieval latency and inter-channel communication overhead, complemented by a dedicated cache that exploits data locality and enhances prefetch efficiency. With these co-optimized techniques, NASZIP delivers speedups of up to $8.4\times$ / $1.4\times$ over CPU baseline and state-of-the-art GPU implementation at equal accuracy. Relative to the state-of-the-art NDP ANNS accelerator ANSMET, NASZIP achieves $1.69\times$ performance improvement.

cs.AR↗

arXiv2026

EPIC: Abstraction and Polymorphism of In-Network Collectives on Ethernet

In-Network Collective (INC) acceleration holds immense potential for optimizing AI training and inference; however, its cross-layer nature has historically hindered investment and adoption within the open Ethernet ecosystem. To bridge this gap, we propose EPIC (Ethernet Polymorphic In-network Collective), an INC protocol specification and reference system built on the principle of "Unified Abstraction, Polymorphic Realization." EPIC introduces an abstraction compatible with standard Ethernet that aligns functional boundaries with participant roles, while offering polymorphic realizations tailored to varying hardware capabilities. We address three fundamental challenges: first, we employ a modular design that enables an evolutionary path from simple to complex implementations, allowing vendors to iterate their hardware incrementally; second, we apply formal verification methodologies to prove the correctness of all proposed polymorphic modes; and third, we develop a unified resource management model versatile enough for diverse INC scenarios. Extensive validation -- spanning model checking, packet/flow simulations, VM emulation, Tofino Testbed, and FPGA/RTL verification -- confirms EPIC's correctness, performance gain, and feasibility.

cs.DC↗

arXiv2026

Multi-element Persuasion in Social Media Health Communication: Synergistic and Trade-off Effects

Health messages on social media are typically constructed through combinations of source cues, appeals, frames, and evidence, which jointly shape communication and persuasive effects. However, prior research has largely focused on single elements or simple pairwise interactions, offering insufficient insight into how multiple elements operate together in real-world digital environments. To address this gap, this study adopts a systems perspective to examine multi-element message combinations. Using 1.8 million health-related Weibo posts, we apply clustering analysis to identify recurring combinations and assess their relationships with communication effects. First, four recurring element combinations are identified: Institutional Authority, Narrative, Assertive Appeal, and Contextual Expression. These combinations function as core structures organized around two key elements. Second, stronger communication effects depend not only on core structures but also on peripheral elements aligned with these structures, with combinations of two to four peripheral elements generally showing greater advantages. Third, the optimal level of peripheral complexity varies with source influence, indicating that environmental factors condition the relationship between message combinations and communication effects. These findings show that communication and persuasive effects are shaped by synergies and trade-offs among multiple persuasive elements. Based on this, the study proposes a Core-Periphery-Environment framework to explain how message combinations generate communication effects with persuasive implications on social media. The study extends research from isolated elements to systems combinations and offers practical implications for health communication.

cs.CY↗

arXiv2026

Gender Bias in YouTube Exposure: Allocative and Structural Inequalities in Political Information Environments

Recommendation algorithms have become the dominant mechanism for information distribution on digital platforms, profoundly shaping personalized information consumption environments. However, gender bias, as a significant form of algorithmic discrimination, may cause users to experience unequal exposure within different political information environments. Taking YouTube as a case, we conduct a controlled social-bot field experiment, where male-coded and female-coded profiles are constructed. We track the exposure and click patterns of these bots to analyze their recommendation trajectories. We analyze the distribution of recommended content from two dimensions: allocative bias and structural bias. First, we find statistically significant differences in allocative bias across male-coded and female-coded profiles, particularly in terms of issue distribution, ideological orientation, and political entities. Secondly, we observe structural bias in the political information environments, characterized by distinct clustering patterns. Additionally, time-series analysis shows that exposure pathways continue to be shaped over time by both communities detected in the co-occurrence network and individual profile-level dynamics. Finally, we construct a simple collaborative-filtering model that reproduces the observed gender bias. We argue that gender bias in recommendation systems is reflected not only in the allocation of political content, but also in how community structures shape these environments, reinforcing societal inequalities and highlighting the need for algorithmic fairness.

cs.SI↗

arXiv2026

SIGMark: Scalable In-Generation Watermark with Blind Extraction for Video Diffusion

Artificial Intelligence Generated Content (AIGC), particularly video generation with diffusion models, has been advanced rapidly. Invisible watermarking is a key technology for protecting AI-generated videos and tracing harmful content, and thus plays a crucial role in AI safety. Beyond post-processing watermarks which inevitably degrade video quality, recent studies have proposed distortion-free in-generation watermarking for video diffusion models. However, existing in-generation approaches are non-blind: they require maintaining all the message-key pairs and performing template-based matching during extraction, which incurs prohibitive computational costs at scale. Moreover, when applied to modern video diffusion models with causal 3D Variational Autoencoders (VAEs), their robustness against temporal disturbance becomes extremely weak. To overcome these challenges, we propose SIGMark, a Scalable In-Generation watermarking framework with blind extraction for video diffusion. To achieve blind-extraction, we propose to generate watermarked initial noise using a Global set of Frame-wise PseudoRandom Coding keys (GF-PRC), reducing the cost of storing large-scale information while preserving noise distribution and diversity for distortion-free watermarking. To enhance robustness, we further design a Segment Group-Ordering module (SGO) tailored to causal 3D VAEs, ensuring robust watermark inversion during extraction under temporal disturbance. Comprehensive experiments on modern diffusion models show that SIGMark achieves very high bit-accuracy during extraction under both temporal and spatial disturbances with minimal overhead, demonstrating its scalability and robustness. Our project is available at https://jeremyzhao1998.github.io/SIGMark-release/.

cs.CV↗

arXiv2026

Studies on the spin and magnetic inclination evolution of magnetars Swift J1834.9-0846 under wind braking

The magnetar Swift J1834.9-0846 presents a significant challenge to neutron star spin-down models. It exhibits two key anomalies: an insufficient rotational energy loss rate to power its observed X-ray luminosity, and a braking index of $ = 1.08\pm 0.04$, which starkly contradicts the canonical magnetic dipole value of $n=3$. To explain these anomalies, we develop a unified spin-evolution model that self-consistently integrates magnetic dipole radiation, gravitational wave emission, and wind braking. Within this framework, we constrain the wind braking parameter to $\kappa \in [13, 37]$ from the nebular properties, finding it contributes substantially (17%-51%) to the current spin-down torque. Bayesian inference reveals that the birth period is poorly constrained by present data and is prior-dependent, indicating a millisecond birth is allowed but not required. Furthermore, we constrain the number of precession cycles to $\xi \sim 10^{4}$--$10^{5}$, and our analysis favors a toroidally-dominated internal magnetic field configuration as the most self-consistent explanation for the low braking index. Finally, we assess the continuous gravitational-wave detectability. The present-day signal is undetectable. However, the early-time signal might have reached the projected sensitivity of next-generation gravitational-wave observatories, such as the Advanced Laser Interferometer Gravitational-Wave Observatory (aLIGO) and the Einstein Telescope (ET), although a confident detection would require exceptionally stable rotation, an assumption considered highly optimistic for a young magnetar. This work establishes a unified framework that links magnetar spin-down with their interior physics and multi-messenger observables, providing a physically consistent interpretation for Swift J1834.9-0846 and a new tool for understanding similar extreme neutron stars.

astro-ph.HE↗

arXiv2025

Exploiting Block Coordinate Descent for Cost-Effective LLM Model Training

Training large language models typically demands extensive GPU memory and substantial financial investment, which poses a barrier for many small- to medium-sized teams. In this paper, we propose a full-parameter pre-training and fine-tuning framework based on block coordinate descent (BCD), enhanced with engineering optimizations, to enable efficient training of large-scale models on cost-effective RTX 4090, A100 and A800 GPU clusters. Under identical hardware configurations, we reduce the training cost of a 7B model to 33% on A100/A800 and only 2.6% on RTX 4090, compared to standard full-parameter training. It also enables large models previously restricted to A100 clusters to be trained on RTX 4090 without degrading performance. BCD achieves comparable or better accuracy than full-parameter and fine-tuning methods at most cases, with lower GPU consumption and improved hardware utilization.

cs.LG↗

arXiv2025

Invariant-based Robust Weights Watermark for Large Language Models

Watermarking technology has gained significant attention due to the increasing importance of intellectual property (IP) rights, particularly with the growing deployment of large language models (LLMs) on billions resource-constrained edge devices. To counter the potential threats of IP theft by malicious users, this paper introduces a robust watermarking scheme without retraining or fine-tuning for transformer models. The scheme generates a unique key for each user and derives a stable watermark value by solving linear constraints constructed from model invariants. Moreover, this technology utilizes noise mechanism to hide watermark locations in multi-user scenarios against collusion attack. This paper evaluates the approach on three popular models (Llama3, Phi3, Gemma), and the experimental results confirm the strong robustness across a range of attack methods (fine-tuning, pruning, quantization, permutation, scaling, reversible matrix and collusion attacks).

cs.CR↗

arXiv2025

A Prior-Guided Joint Diffusion Model in Projection Domain for PET Tracer Conversion

Positron emission tomography (PET) is widely used to assess metabolic activity, but its application is limited by the availability of radiotracers. 18F-labeled fluorodeoxyglucose (18F-FDG) is the most commonly used tracer but shows limited effectiveness for certain tumors. In contrast, 6-18F-fluoro-3,4-dihydroxy-L-phenylalanine (18F-DOPA) offers higher specificity for neuroendocrine tumors and neurological disorders. However, the complexity of its synthesis process and constraints on transportation time have limited its clinical application. Among different forms of raw data acquired by the scanner, sinogram is a commonly used representation in PET imaging. Therefore, modeling in projection domain enables more direct utilization of the original information, potentially reducing the accumulation errors during the image reconstruction process. Inspired by these factors, this study proposes a prior-guided joint diffusion model (PJDM) for transforming 18F-FDG PET sinograms into 18F-DOPA PET sinograms. During inference, an initial synthetic 18F-DOPA PET sinogram is first generated using a higher-order hybrid sampler. This sinogram is then degraded and serves as an additional condition to guide the iterative refinement process. Experimental results demonstrated that PJDM effectively improved both sinogram quality and the final synthetic outcomes. The code is available at: https://github.com/yqx7150/PJDM.

eess.IV↗

arXiv2025

C-LoRA: Contextual Low-Rank Adaptation for Uncertainty Estimation in Large Language Models

Low-Rank Adaptation (LoRA) offers a cost-effective solution for fine-tuning large language models (LLMs), but it often produces overconfident predictions in data-scarce few-shot settings. To address this issue, several classical statistical learning approaches have been repurposed for scalable uncertainty-aware LoRA fine-tuning. However, these approaches neglect how input characteristics affect the predictive uncertainty estimates. To address this limitation, we propose Contextual Low-Rank Adaptation (C-LoRA) as a novel uncertainty-aware and parameter efficient fine-tuning approach, by developing new lightweight LoRA modules contextualized to each input data sample to dynamically adapt uncertainty estimates. Incorporating data-driven contexts into the parameter posteriors, C-LoRA mitigates overfitting, achieves well-calibrated uncertainties, and yields robust predictions. Extensive experiments on LLaMA2-7B models demonstrate that C-LoRA consistently outperforms the state-of-the-art uncertainty-aware LoRA methods in both uncertainty quantification and model generalization. Ablation studies further confirm the critical role of our contextual modules in capturing sample-specific uncertainties. C-LoRA sets a new standard for robust, uncertainty-aware LLM fine-tuning in few-shot regimes. Although our experiments are limited to 7B models, our method is architecture-agnostic and, in principle, applies beyond this scale; studying its scaling to larger models remains an open problem. Our code is available at https://github.com/ahra99/c_lora.

cs.LG↗

arXiv2025

VLM-based Prompts as the Optimal Assistant for Unpaired Histopathology Virtual Staining

In histopathology, tissue sections are typically stained using common H&E staining or special stains (MAS, PAS, PASM, etc.) to clearly visualize specific tissue structures. The rapid advancement of deep learning offers an effective solution for generating virtually stained images, significantly reducing the time and labor costs associated with traditional histochemical staining. However, a new challenge arises in separating the fundamental visual characteristics of tissue sections from the visual differences induced by staining agents. Additionally, virtual staining often overlooks essential pathological knowledge and the physical properties of staining, resulting in only style-level transfer. To address these issues, we introduce, for the first time in virtual staining tasks, a pathological vision-language large model (VLM) as an auxiliary tool. We integrate contrastive learnable prompts, foundational concept anchors for tissue sections, and staining-specific concept anchors to leverage the extensive knowledge of the pathological VLM. This approach is designed to describe, frame, and enhance the direction of virtual staining. Furthermore, we have developed a data augmentation method based on the constraints of the VLM. This method utilizes the VLM's powerful image interpretation capabilities to further integrate image style and structural information, proving beneficial in high-precision pathological diagnostics. Extensive evaluations on publicly available multi-domain unpaired staining datasets demonstrate that our method can generate highly realistic images and enhance the accuracy of downstream tasks, such as glomerular detection and segmentation. Our code is available at: https://github.com/CZZZZZZZZZZZZZZZZZ/VPGAN-HARBOR

eess.IV↗

arXiv2024

Enhancing Visual Representation for Text-based Person Searching

Text-based person search aims to retrieve the matched pedestrians from a large-scale image database according to the text description. The core difficulty of this task is how to extract effective details from pedestrian images and texts, and achieve cross-modal alignment in a common latent space. Prior works adopt image and text encoders pre-trained on unimodal data to extract global and local features from image and text respectively, and then global-local alignment is achieved explicitly. However, these approaches still lack the ability of understanding visual details, and the retrieval accuracy is still limited by identity confusion. In order to alleviate the above problems, we rethink the importance of visual features for text-based person search, and propose VFE-TPS, a Visual Feature Enhanced Text-based Person Search model. It introduces a pre-trained multimodal backbone CLIP to learn basic multimodal features and constructs Text Guided Masked Image Modeling task to enhance the model's ability of learning local visual details without explicit annotation. In addition, we design Identity Supervised Global Visual Feature Calibration task to guide the model learn identity-aware global visual features. The key finding of our study is that, with the help of our proposed auxiliary tasks, the knowledge embedded in the pre-trained CLIP model can be successfully adapted to text-based person search task, and the model's visual understanding ability is significantly enhanced. Experimental results on three benchmarks demonstrate that our proposed model exceeds the existing approaches, and the Rank-1 accuracy is significantly improved with a notable margin of about $1\%\sim9\%$. Our code can be found at https://github.com/zhangweifeng1218/VFE_TPS.

cs.CV↗

arXiv2024

High-density single-atom electrocatalytic centers on two-dimensional topological platinum tellurides with Te-vacancy superstructure

Chemical activation of the intrinsically inert basal planes of transition metal dichalcogenides (TMDs) is crucial for developing high-efficiency electrocatalysts for energy technology applications. Here we report the discovery of an efficient TMD-based topological catalyst for hydrogen evolution reaction (HER), containing high-density single-atom reactive centers on a few-layer (7x7)-PtTe2-x superstructure with a Te-vacancy density of x. Compared with pristine Pt(111), PtTe2, and (2x2)-PtTe2-x, (7x7)-PtTe2-x exhibits superior HER performance owing to its substantially increased density of undercoordinated Pt sites, and displays exceptional catalytic stability when operating at high current densities. Our first-principles calculations confirm that multiple types of undercoordinated Pt sites on (7x7)-PtTe2-x exhibit favorable hydrogen adsorption Gibbs free energies, and that the reactive sites can further increase their population upon increasing hydrogen coverage. Both the (2x2)- and (7x7)-PtTe2-x are also shown to possess nontrivial band topology with robust edge states that may further facilitate HER.

cond-mat.mtrl-sci↗

arXiv2024

Arena: Efficiently Training Large Models via Dynamic Scheduling and Adaptive Parallelism Co-Design

Efficiently training large-scale models (LMs) in GPU clusters involves two separate avenues: inter-job dynamic scheduling and intra-job adaptive parallelism (AP). However, existing dynamic schedulers struggle with large-model scheduling due to the mismatch between static parallelism (SP)-aware scheduling and AP-based execution, leading to cluster inefficiencies such as degraded throughput and prolonged job queuing. This paper presents Arena, a large-model training system that co-designs dynamic scheduling and adaptive parallelism to achieve high cluster efficiency. To reduce scheduling costs while improving decision quality, Arena designs low-cost, disaggregated profiling and AP-tailored, load-aware performance estimation, while unifying them by sharding the joint scheduling-parallelism optimization space via a grid abstraction. Building on this, Arena dynamically schedules profiled jobs in elasticity and heterogeneity dimensions, and executes them using efficient AP with pruned search space. Evaluated on heterogeneous testbeds and production workloads, Arena reduces job completion time by up to $49.3\%$ and improves cluster throughput by up to $1.60\times$.

cs.DC↗

arXiv2024

Challenges and Opportunities to Enable Large-Scale Computing via Heterogeneous Chiplets

Fast-evolving artificial intelligence (AI) algorithms such as large language models have been driving the ever-increasing computing demands in today's data centers. Heterogeneous computing with domain-specific architectures (DSAs) brings many opportunities when scaling up and scaling out the computing system. In particular, heterogeneous chiplet architecture is favored to keep scaling up and scaling out the system as well as to reduce the design complexity and the cost stemming from the traditional monolithic chip design. However, how to interconnect computing resources and orchestrate heterogeneous chiplets is the key to success. In this paper, we first discuss the diversity and evolving demands of different AI workloads. We discuss how chiplet brings better cost efficiency and shorter time to market. Then we discuss the challenges in establishing chiplet interface standards, packaging, and security issues. We further discuss the software programming challenges in chiplet systems.

cs.AR↗

arXiv2023

High-throughput optical neural networks based on temporal computing

An emerging generative artificial intelligence (AI) based on neural networks starts to grow in popularity with a revolutionizing capability of creating new and original content. As giant generative models with millions to billions of parameters are developed, trained and maintained, a massive and energy-efficient computing power is highly required. However, conventional digital computers are struggling to keep up with the pace of the generative model improvements. In this paper, we propose and demonstrate high-throughput optical neural networks based on temporal computing. The core weighted summation operation is realized with the use of high-speed electro-optic modulation and low-speed balanced photodetection. The input data and weight are encoded in a time sequence separately and loaded on an optical signal via two electro-optic modulators sequentially. By precisely controlling the synchronization time of the data and weight loading, the matrix multiplication is performed. Followed by a balanced photodetector, the summation is conducted, thanks to the electron accumulation of the inherent electronic integrator circuit of the low-speed photodetector. Thus, the linear weighted summation operation is implemented based on temporal computing in the optical domain. With the proposed optical linear weighted summation, a fully-connected neural network and convolutional neural network are realized. Thanks to the high-speed feature of temporal computing, a high data throughput of the optical neural network is experimentally demonstrated, and the weighting coefficients can be specified on demand, which enables a strong programmability of the optical neural network. By leveraging wavelength multiplexing technology, a scalable optical neural network could be created with a massive computing power and strong reconfigurability, which holds great potential for future giant AI applications.

cs.ET↗

arXiv2023

Microdisk modulator-assisted optical nonlinear activation functions for photonic neural networks

On-chip implementation of optical nonlinear activation functions (NAFs) is essential for realizing large-scale photonic neural chips. To implement different neural processing and machine learning tasks with optimal performances, different NAFs are explored with the use of different devices. From the perspective of on-chip integration and reconfigurability of photonic neural network (PNN), it is highly preferred that a single compact device can fulfill multiple NAFs. Here, we propose and experimentally demonstrate a compact high-speed microdisk modulator to realize multiple NAFs. The fabricated microdisk modulator has an add-drop configuration in which a lateral PN junction is incorporated for tuning. Based on high-speed nonlinear electrical-optical (E-O) effect, multiple NAFs are realized by electrically controlling free-carrier injection. Thanks to its strong optical confinement of the disk cavity, all-optical thermo-optic (TO) nonlinear effect can also be leveraged to realize other four different NAFs, which is difficult to be realized with the use of electrical-optical effect. With the use of the realized nonlinear activation function, a convolutional neural network (CNN) is studied to perform handwritten digit classification task, and an accuracy as large as 98% is demonstrated, which verifies the effectiveness of the use of the high-speed microdisk modulator to realize the NAFs. Thanks to its compact footprint and strong electrical-optical or all-optical effects, the microdisk modulator features multiple NAFs, which could serve as a flexible nonlinear unit for large-scale PNNs.

physics.optics↗