arXiv subjects

Zhibin Zhang

Publications and source records attributed to Zhibin Zhang.

17 recordsLinked to original sources

arXiv2026

RepoOMP: Repository-Aware Hotspot OpenMP Parallelization via Dependency-Aware Context Reduction

OpenMP parallelization of hotspots in mature repositories remains difficult because loop safety and optimization payoff often depend on non-local evidence. Rule-based tools under-parallelize when legality is not locally provable, while agent-based approaches become unstable when retrieval misses decisive dependencies or includes irrelevant code. We present RepoOMP, a hybrid framework that recovers parallelization-relevant evidence before generation. RepoOMP builds a Multi-granularity Attributes Performance graph (MAP), routes hotspots between deterministic rules and an LLM agent, and constructs a Structured Transformation Context (STC) that exposes dependency facts without flooding the model with unrelated repository text. We evaluate RepoOMP on 951 profiled hotspots from NPB, BOTS, FFmpeg, NCNN, and GROMACS. Under compilation, workload-specific checks, and positive speedup, 372 hotspots are accepted, including 330 real-world repository hotspots. RepoOMP achieves average speedups of $8.23\times$ on NPB and $8.96\times$ on BOTS. For the nine detailed real-world kernels used in matched-backbone and robustness analyses, RepoOMP reaches a cross-backbone mean of $5.25\times$, improves speedup by 18--28\%, and reduces agent-side token cost by 47--68\% relative to the unstructured Claude Code baseline. Across 330 accepted real-world hotspots, median speedup is $2.25\times$. Overall, RepoOMP provides an evidence-guided workflow for hotspot parallelization in repository settings. The open-source repository is available at https://github.com/Qlalq/RepoOMP_Simplified.

cs.DC↗

arXiv2026

ReFLEX: Length-Generalizable CSI Denoising for MIMO-OFDM via Relative-Frequency Bias

This letter studies CSI denoising for MIMO--OFDM with variable NR resource block (RB) allocations. ReFLEX is a length-generalizable Transformer whose frequency attention uses a relative-frequency position bias (RFPB) generated from subcarrier offsets. A single checkpoint handles unseen RB lengths and can be applied to sparse DM-RS observations in the tested RB5/RB10 PUSCH setup without retraining. In a 3GPP~TR~38.901 UMa NLOS channel, ReFLEX achieves about $-9.6$~dB NMSE on unseen RB lengths. In NR PUSCH/UL-SCH simulations, ReFLEX denoising followed by time-frequency interpolation reduces the 10\% BLER threshold by about 2--3~dB.

eess.SP↗

arXiv2026

EGAD: Entropy-Guided Adaptive Distillation for Token-Level Knowledge Transfer

Large language models (LLMs) have achieved remarkable performance across diverse domains, yet their enormous computational and memory requirements hinder deployment in resource-constrained environments. Knowledge distillation offers a promising solution by transferring knowledge from a large teacher model to a smaller student model. However, existing distillation methods typically treat all tokens equally, ignoring the fact that different tokens contribute unequally to model decisions. This can lead to inefficient knowledge transfer and reduced learning effectiveness. To address this limitation, we propose an entropy-based adaptive distillation strategy that dynamically adjusts the training process at the token level. Our method leverages the teacher's output entropy to guide three aspects of distillation. Specifically, we introduce a token-level curriculum by dynamically shifting focus from low- to high-entropy tokens during training. We further adjust the distillation temperature based on token entropy to better capture teacher confidence patterns. Moreover, we employ a dual-branch architecture for efficient logits-only distillation on easy tokens and deeper feature-based distillation on difficult tokens. Extensive experiments validate the soundness and effectiveness of our method.

cs.CL↗

arXiv2026

MI-PRUN: Optimize Large Language Model Pruning via Mutual Information

Large Language Models (LLMs) have become indispensable across various domains, but this comes at the cost of substantial computational and memory resources. Model pruning addresses this by removing redundant components from models. In particular, block pruning can achieve significant compression and inference acceleration. However, existing block pruning methods are often unstable and struggle to attain globally optimal solutions. In this paper, we propose a mutual information based pruning method MI-PRUN for LLMs. Specifically, we leverages mutual information to identify redundant blocks by evaluating transitions in hidden states. Additionally, we incorporate the Data Processing Inequality (DPI) to reveal the relationship between the importance of entire contiguous blocks and that of individual blocks. Moreover, we develop the Fast-Block-Select algorithm, which iteratively updates block combinations to achieve a globally optimal solution while significantly improving the efficiency. Extensive experiments across various models and datasets demonstrate the stability and effectiveness of our method.

cs.CL↗

arXiv2025

MoRA: On-the-fly Molecule-aware Low-Rank Adaptation Framework for LLM-based Multi-Modal Molecular Assistant

Effectively integrating molecular graph structures with Large Language Models (LLMs) is a key challenge in drug discovery. Most existing multi-modal alignment methods typically process these structures by fine-tuning the LLM or adding a static adapter simultaneously. However, these approaches have two main limitations: (1) it optimizes a shared parameter space across all molecular inputs, limiting the model's ability to capture instance-specific structural features; and (2) fine-tuning the LLM for molecular tasks can lead to catastrophic forgetting, undermining its general reasoning capabilities. In this paper, instead of static task-oriented adaptation, we propose an instance-specific parameter space alignment approach for each molecule on-the-fly. To this end, we introduce Molecule-aware Low-Rank Adaptation (MoRA) that produces a unique set of low-rank adaptation weights for each input molecular graph. These weights are then dynamically injected into a frozen LLM, allowing the model to adapt its reasoning to the structure of each molecular input, while preserving the LLM's core knowledge. Extensive experiments demonstrate that on key molecular tasks, such as chemical reaction prediction and molecular captioning, MoRA's instance-specific dynamic adaptation outperforms statically adapted baselines, including a 14.1% relative improvement in reaction prediction exact match and a 22% reduction in error for quantum property prediction. The code is available at https://github.com/jk-sounds/MoRA.

cs.LG↗

arXiv2025

ScatterAD: Temporal-Topological Scattering Mechanism for Time Series Anomaly Detection

One main challenge in time series anomaly detection for industrial IoT lies in the complex spatio-temporal couplings within multivariate data. However, traditional anomaly detection methods focus on modeling spatial or temporal dependencies independently, resulting in suboptimal representation learning and limited sensitivity to anomalous dispersion in high-dimensional spaces. In this work, we conduct an empirical analysis showing that both normal and anomalous samples tend to scatter in high-dimensional space, especially anomalous samples are markedly more dispersed. We formalize this dispersion phenomenon as scattering, quantified by the mean pairwise distance among sample representations, and leverage it as an inductive signal to enhance spatio-temporal anomaly detection. Technically, we propose ScatterAD to model representation scattering across temporal and topological dimensions. ScatterAD incorporates a topological encoder for capturing graph-structured scattering and a temporal encoder for constraining over-scattering through mean squared error minimization between neighboring time steps. We introduce a contrastive fusion mechanism to ensure the complementarity of the learned temporal and topological representations. Additionally, we theoretically show that maximizing the conditional mutual information between temporal and topological views improves cross-view consistency and enhances more discriminative representations. Extensive experiments on multiple public benchmarks show that ScatterAD achieves state-of-the-art performance on multivariate time series anomaly detection. Code is available at this repository: https://github.com/jk-sounds/ScatterAD.

cs.LG↗

arXiv2025

SpeedLLM: An FPGA Co-design of Large Language Model Inference Accelerator

This paper introduces SpeedLLM, a neural network accelerator designed on the Xilinx Alevo U280 platform and optimized for the Tinyllama framework to enhance edge computing performance. Key innovations include data stream parallelism, a memory reuse strategy, and Llama2 operator fusion, which collectively reduce latency and energy consumption. SpeedLLM's data pipeline architecture optimizes the read-compute-write cycle, while the memory strategy minimizes FPGA resource demands. The operator fusion boosts computational density and throughput. Results show SpeedLLM outperforms traditional Tinyllama implementations, achieving up to 4.8* faster performance and 1.18* lower energy consumption, offering improvements in edge devices.

cs.AR↗

arXiv2024

Towards Cross-domain Few-shot Graph Anomaly Detection

Few-shot graph anomaly detection (GAD) has recently garnered increasing attention, which aims to discern anomalous patterns among abundant unlabeled test nodes under the guidance of a limited number of labeled training nodes. Existing few-shot GAD approaches typically adopt meta-training methods trained on richly labeled auxiliary networks to facilitate rapid adaptation to target networks that possess sparse labels. However, these proposed methods often assume that the auxiliary and target networks exist in the same data distributions-an assumption rarely holds in practical settings. This paper explores a more prevalent and complex scenario of cross-domain few-shot GAD, where the goal is to identify anomalies within sparsely labeled target graphs using auxiliary graphs from a related, yet distinct domain. The challenge here is nontrivial owing to inherent data distribution discrepancies between the source and target domains, compounded by the uncertainties of sparse labeling in the target domain. In this paper, we propose a simple and effective framework, termed CDFS-GAD, specifically designed to tackle the aforementioned challenges. CDFS-GAD first introduces a domain-adaptive graph contrastive learning module, which is aimed at enhancing cross-domain feature alignment. Then, a prompt tuning module is further designed to extract domain-specific features tailored to each domain. Moreover, a domain-adaptive hypersphere classification loss is proposed to enhance the discrimination between normal and anomalous instances under minimal supervision, utilizing domain-sensitive norms. Lastly, a self-training strategy is introduced to further refine the predicted scores, enhancing its reliability in few-shot settings. Extensive experiments on twelve real-world cross-domain data pairs demonstrate the effectiveness of the proposed CDFS-GAD framework in comparison to various existing GAD methods.

cs.LG↗

arXiv2024

Phonon heat conduction across slippery interfaces in twisted graphite

Interlayer rotation in van der Waals (vdW) materials offers great potential for manipulating phonon dynamics and heat flow in advanced electronics with ever higher compactness and power density. However, despite extensive theoretical efforts in recent years, experimental measurements remain scarce especially due to the critical challenges of preparing single-crystalline twisted interfaces and probing interfacial thermal transport with sufficient resolution. Here, we exploited the intrinsic twisted interfaces in highly oriented pyrolytic graphite (HOPG). By developing novel experimental schemes based on microfabricated mesas, we managed to achieve simultaneous mechanical characterizations and thermal measurements. In particular, we pushed the HOPG mesas with a microprobe to identify and rotate single-crystalline intrinsic interfaces owing to their slippery nature as is well known in structural superlubricity. Remarkably, we observed over 30-fold suppression of thermal conductance for the slippery interfaces by using epitaxial graphite as a control. Nonetheless, the interfacial conductance remains around 600 $\mathrm{MWm^{-2}K^{-1}}$ which surpasses the highest values for artificially stacked vdW structures by more than five times. Further, atomic simulations revealed the predominant role of the transverse acoustic phonons. Together, our findings highlight a general physical picture that directly correlates interfacial thermal transport with sliding resistance, and lay the foundation for twist-enabled thermal management which are particularly beneficial to twistronics and slidetronics.

cond-mat.mes-hall↗

arXiv2023

Growth and applications of two-dimensional single crystals

Two-dimensional (2D) materials have received extensive research attentions over the past two decades due to their intriguing physical properties (such as the ultrahigh mobility and strong light-matter interaction at atomic thickness) and a broad range of potential applications (especially in the fields of electronics and optoelectronics). The growth of single-crystal 2D materials is the prerequisite to realize 2D-based high-performance applications. In this review, we aim to provide an in-depth analysis of the state-of-the-art technology for the growth and applications of 2D materials, with particular emphasis on single crystals. We first summarize the major growth strategies for monolayer 2D single crystals. Following that, we discuss the growth of multilayer single crystals, including the control of thickness, stacking sequence, and heterostructure composition. Then we highlight the exploration of 2D single crystals in electronic and optoelectronic devices. Finally, a perspective is given to outline the research opportunities and the remaining challenges in this field.

cond-mat.mtrl-sci↗

arXiv2023

GEDI: A Graph-based End-to-end Data Imputation Framework

Data imputation is an effective way to handle missing data, which is common in practical applications. In this study, we propose and test a novel data imputation process that achieve two important goals: (1) preserve the row-wise similarities among observations and column-wise contextual relationships among features in the feature matrix, and (2) tailor the imputation process to specific downstream label prediction task. The proposed imputation process uses Transformer network and graph structure learning to iteratively refine the contextual relationships among features and similarities among observations. Moreover, it uses a meta-learning framework to select features that are influential to the downstream prediction task of interest. We conduct experiments on real-world large data sets, and show that the proposed imputation process consistently improves imputation and label prediction performance over a variety of benchmark methods.

cs.LG↗

arXiv2023

Graphene/silicon heterojunction for reconfigurable phase-relevant activation function in coherent optical neural networks

Optical neural networks (ONNs) herald a new era in information and communication technologies and have implemented various intelligent applications. In an ONN, the activation function (AF) is a crucial component determining the network performances and on-chip AF devices are still in development. Here, we first demonstrate on-chip reconfigurable AF devices with phase activation fulfilled by dual-functional graphene/silicon (Gra/Si) heterojunctions. With optical modulation and detection in one device, time delays are shorter, energy consumption is lower, reconfigurability is higher and the device footprint is smaller than other on-chip AF strategies. The experimental modulation voltage (power) of our Gra/Si heterojunction achieves as low as 1 V (0.5 mW), superior to many pure silicon counterparts. In the photodetection aspect, a high responsivity of over 200 mA/W is realized. Special nonlinear functions generated are fed into a complex-valued ONN to challenge handwritten letters and image recognition tasks, showing improved accuracy and potential of high-efficient, all-component-integration on-chip ONN. Our results offer new insights for on-chip ONN devices and pave the way to high-performance integrated optoelectronic computing circuits.

physics.optics↗

arXiv2022

The Statistical Similarity of Repeating and Non-Repeating Fast Radio Bursts

In this paper, we present a sample of 21 repeating fast radio bursts (FRBs) detected by different radio instruments before September 2021. Using the Anderson--Darling test, we compared the distributions of extra-Galactic dispersion measure ($DM_{\rm E}$) of non-repeating FRBs, repeating FRBs and all FRBs. It was found that the $ DM_{\rm E}$ values of three sub-samples are log-normally distributed. The $DM_{\rm E}$ of repeaters and non-repeaters were drawn from a different distribution on basis of the Mann--Whitney--Wilcoxon test. In addition, assuming that the non-repeating FRBs identified currently may be potentially repeators, i.e., the repeating FRBs to be universal and representative, one can utilize the averaged fluence of repeating FRBs as an indication from which to derive an apparent intensity distribution function (IDF) with a power-law index of $a_1=$ $1.10\pm 0.14$ ($a_2=$ $1.01\pm 0.16$, the observed fluence as a statistical variant), which is in good agreement with the previous IDF of 16 non-repeating FRBs found by Li et al. Based on the above statistics of repeating and non-repeating FRBs, we propose that both types of FRBs may have different cosmological origins, spatial distributions and circum-burst environments. Interestingly, the differential luminosity distributions of repeating and non-repeating FRBs can also be well described by a broken power-law function with the same power-law index of $-$1.4.

astro-ph.HE↗

arXiv2019

Kaleido: An Efficient Out-of-core Graph Mining System on A Single Machine

Graph mining is one of the most important categories of graph algorithms. However, exploring the subgraphs of an input graph produces a huge amount of intermediate data. The 'think like a vertex' programming paradigm, pioneered by Pregel, cannot readily formulate mining problems, which is designed to produce graph computation problems like PageRank. Existing mining systems like Arabesque and RStream need large amounts of computing and memory resources. In this paper, we present Kaleido, an efficient single machine, out-of-core graph mining system which treats disks as an extension of memory. Kaleido treats intermediate data in graph mining tasks as a tensor and adopts a succinct data structure for the intermediate data. Kaleido utilizes the eigenvalue of the adjacency matrix of a subgraph to efficiently solve the subgraph isomorphism problems with an acceptable constraint that the vertex number of a subgraph is less than 9. Kaleido implements half-memory-half-disk storage for storing large intermediate data, which treats the disk as an extension of the memory. Comparing with two state-of-the-art mining systems, Arabesque and RStream, Kaleido outperforms them by a GeoMean 12.3$\times$ and 40.0$\times$ respectively.

cs.DC↗

arXiv2015

Order and Chaos near Equilibrium Points in the Potential of Rotating Highly Irregular-shaped Celestial Bodies

The order and chaos of the motion near equilibrium points in the potential of a rotating highly irregular-shaped celestial body are investigated from point of view of the dynamical system theory. The positions of the non-degenerate equilibrium points vary continuously when the parameter changes. The topological structures in the vicinity of equilibrium points are classified into several different cases. Bifurcations at equilibrium points and the topological transfers between different cases for equilibrium points are also discussed. The conclusions can be applied to all kinds of rotating celestial bodies, simple-shaped or highly irregular-shaped, including asteroids, comets, planets and satellites of planets to help one to understand the dynamical behaviors around them. Applications to asteroids 216 Kleopatra, 2063 Bacchus, and 25143 Itokawa are significant and interesting: eigenvalues affiliated to the equilibrium points for the asteroid 216 Kleopatra move and always belong to the same topological cases; while eigenvalues affiliated to two different equilibrium points for the asteroid 2063 Bacchus and 25143 Itokawa move through the resonant cases of equilibrium points, and the collision of eigenvalues in the complex plane occurs. Poincaré sections in the potential of the asteroid 216 Kleopatra show the chaos behaviors of the orbits in large scale.

astro-ph.EP↗

arXiv2006

Inferring the emission regions for different kinds of gamma-ray bursts

Using a theoretical model describing pulse shapes, we have clarified the relations between the observed pulses and their corresponding timescales, such as the angular spreading time, the dynamic time as well as the cooling time. We find that the angular spreading timescale caused by curvature effect of fireball surface only contributes to the falling part of the observed pulses, while the dynamic one in the co-moving frame of the shell merely contributes to the rising portion of pulses provided the radiative time is negligible. In addition, the pulses resulted from the pure radiative cooling time of relativistic electrons exhibit properties of fast rise and slow decay (a quasi-FRED) profile together with smooth peaks. Besides, we interpret the phenomena of wider pules tending to be more asymmetric to be a consequence of the difference in emission regions. Meanwhile, we find the intrinsic emission time is decided by the ratios of lorentz factors and radii of the shells between short and long bursts. Based on the analysis of asymmetry, our results suggest that the long GRB pulses may occur in the regions with larger radius, while the short bursts could locate at the smaller distance from central engine.

astro-ph↗

arXiv2006

Revisiting the characteristics of the spectral lags in short gamma-ray bursts

In this paper, we restudy the spectral lag features of short bright gamma-ray bursts (T90 < 2.6s) with a BATSE time-tagged event (TTE) sample including 65 single pulse bursts. We also make an investigation on the characteristics of ratios between the spectral lag and the full width at half maximum ($FWHM$) of the pulses, called relative spectral lags (RSLs). We draw the conclusions as follows: 1) Spectral lags of short GRBs are normally distributed and concentrated on around the value of 0.014 with 40 percent of them having negative lags. With K-S test, we find the lag distribution is identical with a normal one caused by white noises, which indicates the lags of the vast majority of short bursts are so small that they are negligible as Norris et al. have suggested.

astro-ph↗