arXiv subjects

Xiaoxiao Liu

Publications and source records attributed to Xiaoxiao Liu.

10 recordsLinked to original sources

arXiv2026

A minimizing-movement framework for geometric gradient flows with admissible tangential motion

We develop a minimizing-movement framework for parametric finite element approximations of geometric gradient flows with admissible tangential motion. At each time step, the discrete variational problem combines a metric dissipation term for the normal displacement with a surface Dirichlet energy. The metric determines the normal geometric evolution: the $L^2(Γ)$ metric gives mean curvature flow, while the $H^{-1}(Γ)$ metric gives surface diffusion flow. Tangential velocity is selected independently through weak constraints on the deformation map. The central structural condition is admissibility, namely, that the identity map satisfies the constraint. This condition keeps the identity map available as a comparison function and yields the natural stability estimate. The framework recovers the classical Barrett--Garcke--Nürnberg (BGN) scheme from the unconstrained formulation and the dual minimal-deformation-rate (MDR) scheme from the MDR constraint. We further introduce two new admissible variants: an admissible BGN scheme and a relaxed MDR scheme. For the resulting fully discrete schemes, we prove existence and uniqueness under natural nondegeneracy assumptions and establish unconditional energy stability. Numerical experiments compare the admissible and classical schemes and illustrate their stability properties and mesh-quality behavior.

math.NA↗

arXiv2026

Geometric structure-preserving parametric finite element approximations for the constrained Helfrich flow

We propose a structure-preserving parametric finite element method for the constrained Helfrich flow of closed curves and surfaces. The proposed method is based on a two-stage velocity-splitting strategy. In the first stage, the normal velocity is computed from a curvature evolution equation, with the volume and surface area constraints imposed softly in terms of the normal velocity. This step approximates the gradient-flow structure of the Helfrich flow, and its fully discrete parametric finite element approximation leads to a linear system and yields an unconditional energy dissipation estimate at the fully discrete level. In the second stage, the surface mesh is updated by combining the computed normal velocity with a BGN-type tangential velocity. A time-weighted interface normal and an area-correction multiplier are also used to enforce exact preservation of the enclosed volume and surface area. This correction step leads to a nonlinear system, which can be efficiently solved by Newton iteration. The resulting method simultaneously achieves energy decay, exact geometric conservation, and good mesh quality. Numerical experiments for two-dimensional curves and three-dimensional surfaces, including nonsmooth initial data and nonzero spontaneous curvature, are presented to demonstrate the accuracy, robustness, and structure-preserving properties of the proposed method.

math.NA↗

arXiv2026

Do LLMs Triage Like Clinicians? A Dynamic Study of Outpatient Referral

Outpatient referral (OR) is a core clinical workflow that assigns patients to hospital departments under incomplete and evolving information, yet it is commonly simplified as a static classification problem despite being inherently interactive in practice. In this work, we study outpatient referral as a dynamic process driven by information acquisition and uncertainty reduction. We analyze both static scenarios based on fixed patient information and dynamic scenarios involving multi-turn dialogue, to test whether large language models (LLMs) improve referral outcomes through better prediction or more effective questioning. Our findings show that LLMs offer limited advantages over traditional classifiers in static referral accuracy, but consistently outperform them in dynamic settings by asking discriminative follow-up questions that reduce uncertainty over candidate departments. These results suggest that the primary value of LLMs in outpatient referral lies not in static prediction, but in supporting interactive, uncertainty-aware clinical decision-making.

cs.CL↗

arXiv2026

Human or LLM as Standardized Patients? A Comparative Study for Medical Education

Standardized patients (SPs) are indispensable for clinical skills training but remain expensive and difficult to scale. Although large language model (LLM)-based virtual standardized patients (VSPs) have been proposed as an alternative, their behavior remains unstable and lacks rigorous comparison with human standardized patients. We propose EasyMED, a multi-agent VSP framework that separates case-grounded information disclosure from response generation to support stable, inquiry-conditioned patient behavior. We also introduce SPBench, a human-grounded benchmark with eight expert-defined criteria for interaction-level evaluation. Experiments show that EasyMED more closely matches human SP behavior than existing VSPs, particularly in case consistency and controlled disclosure. A four-week controlled study further demonstrates learning outcomes comparable to human SP training, with stronger early gains for novice learners and improved flexibility, psychological safety, and cost efficiency.

cs.CL↗

arXiv2025

Ecco: Improving Memory Bandwidth and Capacity for LLMs via Entropy-aware Cache Compression

Large language models (LLMs) have demonstrated transformative capabilities across diverse artificial intelligence applications, yet their deployment is hindered by substantial memory and computational demands, especially in resource-constrained environments. Quantization techniques have emerged as a critical solution, reducing data precision to enhance memory and computational efficiency. However, existing methods often suffer from high runtime overheads and potential accuracy degradation. To address these challenges, we propose Ecco, an entropy-based cache compression technique tailored for LLMs. Ecco combines group-wise and non-uniform quantization with pre-defined shared k-means patterns and Huffman coding to exploit the inherent entropy characteristics of LLM cache data. Recognizing the inefficiencies of traditional Huffman coding in terms of parallelism and latency, we introduce a novel parallel Huffman-based decoding process with a multi-stage pipeline design, reducing latency by two orders of magnitude and achieving throughput comparable to GPU L2 caches. Comprehensive evaluations demonstrate that Ecco achieves an up to 2.9$\times$ and 1.9$\times$ speedup over the state-of-the-art AWQ and SmoothQuant framework, 2.4$\times$ over the Olive accelerator, all while increasing memory capacity by nearly 4$\times$ and maintaining state-of-the-art LLM accuracy. These results underscore the effectiveness of our entropy-based cache compression in enhancing LLM performance and efficiency, paving the way for more deployable large-scale AI models.

cs.AR↗

arXiv2024

LTRDetector: Exploring Long-Term Relationship for Advanced Persistent Threats Detection

Advanced Persistent Threat (APT) is challenging to detect due to prolonged duration, infrequent occurrence, and adept concealment techniques. Existing approaches primarily concentrate on the observable traits of attack behaviors, neglecting the intricate relationships formed throughout the persistent attack lifecycle. Thus, we present an innovative APT detection framework named LTRDetector, implementing an end-to-end holistic operation. LTRDetector employs an innovative graph embedding technique to retain comprehensive contextual information, then derives long-term features from these embedded provenance graphs. During the process, we compress the data of the system provenance graph for effective feature learning. Furthermore, in order to detect attacks conducted by using zero-day exploits, we captured the system's regular behavior and detects abnormal activities without relying on predefined attack signatures. We also conducted extensive evaluations using five prominent datasets, the efficacy evaluation of which underscores the superiority of LTRDetector compared to existing state-of-the-art techniques.

cs.CR↗

arXiv2022

Reversed Image Signal Processing and RAW Reconstruction. AIM 2022 Challenge Report

Cameras capture sensor RAW images and transform them into pleasant RGB images, suitable for the human eyes, using their integrated Image Signal Processor (ISP). Numerous low-level vision tasks operate in the RAW domain (e.g. image denoising, white balance) due to its linear relationship with the scene irradiance, wide-range of information at 12bits, and sensor designs. Despite this, RAW image datasets are scarce and more expensive to collect than the already large and public RGB datasets. This paper introduces the AIM 2022 Challenge on Reversed Image Signal Processing and RAW Reconstruction. We aim to recover raw sensor images from the corresponding RGBs without metadata and, by doing this, "reverse" the ISP transformation. The proposed methods and benchmark establish the state-of-the-art for this low-level vision inverse problem, and generating realistic raw sensor readings can potentially benefit other tasks such as denoising and super-resolution.

eess.IV↗

arXiv2022

NTIRE 2022 Challenge on Efficient Super-Resolution: Methods and Results

This paper reviews the NTIRE 2022 challenge on efficient single image super-resolution with focus on the proposed solutions and results. The task of the challenge was to super-resolve an input image with a magnification factor of $\times$4 based on pairs of low and corresponding high resolution images. The aim was to design a network for single image super-resolution that achieved improvement of efficiency measured according to several metrics including runtime, parameters, FLOPs, activations, and memory consumption while at least maintaining the PSNR of 29.00dB on DIV2K validation set. IMDN is set as the baseline for efficiency measurement. The challenge had 3 tracks including the main track (runtime), sub-track one (model complexity), and sub-track two (overall performance). In the main track, the practical runtime performance of the submissions was evaluated. The rank of the teams were determined directly by the absolute value of the average runtime on the validation set and test set. In sub-track one, the number of parameters and FLOPs were considered. And the individual rankings of the two metrics were summed up to determine a final ranking in this track. In sub-track two, all of the five metrics mentioned in the description of the challenge including runtime, parameter count, FLOPs, activations, and memory consumption were considered. Similar to sub-track one, the rankings of five metrics were summed up to determine a final ranking. The challenge had 303 registered participants, and 43 teams made valid submissions. They gauge the state-of-the-art in efficient single image super-resolution.

cs.CV↗

arXiv2021

Visualization Resources: A Starting Point

Visualization, as a vibrant field for researchers, practitioners, and higher educational institutions, is growing and evolving very rapidly. Tremendous progress has been made since 1987, the year often cited as the beginning of data visualization as a distinct field. As such, the number of visualization resources and the demand for those resources are increasing at a very fast pace. We present a collection of open visualization resources for all those with an interest in interactive data visualization and visual analytics. Because the number of resources is so large, we focus on collections of resources, of which there are already very many ranging from literature collections to collections of practitioner resources. We develop a novel classification of visualization resource collections based on the resource type, e.g. literature-based, web-based, etc. The result is a helpful overview and details-on-demand of many useful resources. The collection offers a valuable jump-start for those seeking out data visualization resources from all backgrounds spanning from beginners such as students to teachers, practitioners, and researchers wishing to create their own advanced or novel visual designs.

cs.HC↗

arXiv2019

Thread Batching for High-performance Energy-efficient GPU Memory Design

Massive multi-threading in GPU imposes tremendous pressure on memory subsystems. Due to rapid growth in thread-level parallelism of GPU and slowly improved peak memory bandwidth, the memory becomes a bottleneck of GPU's performance and energy efficiency. In this work, we propose an integrated architectural scheme to optimize the memory accesses and therefore boost the performance and energy efficiency of GPU. Firstly, we propose a thread batch enabled memory partitioning (TEMP) to improve GPU memory access parallelism. In particular, TEMP groups multiple thread blocks that share the same set of pages into a thread batch and applies a page coloring mechanism to bound each stream multiprocessor (SM) to the dedicated memory banks. After that, TEMP dispatches the thread batch to an SM to ensure high-parallel memory-access streaming from the different thread blocks. Secondly, a thread batch-aware scheduling (TBAS) scheme is introduced to improve the GPU memory access locality and to reduce the contention on memory controllers and interconnection networks. Experimental results show that the integration of TEMP and TBAS can achieve up to 10.3% performance improvement and 11.3% DRAM energy reduction across diverse GPU applications. We also evaluate the performance interference of the mixed CPU+GPU workloads when they are run on a heterogeneous system that employs our proposed schemes. Our results show that a simple solution can effectively ensure the efficient execution of both GPU and CPU applications.

cs.AR↗