Universal visual anomaly detection aims to identify anomalies from novel or unseen vision domains without additional fine-tuning, which is critical in open scenarios. Recent studies have demonstrated that pre-trained vision-language models like CLIP exhibit strong generalization with just zero or a few normal images. However, existing methods struggle to design prompt templates, handle complex token interactions, or require fine-tuning on target domains, resulting in limited flexibility. In this work, we present a simple yet effective AdaptCLIP based on two key insights. First, adaptive visual and textual representations should be learned alternately rather than jointly. Second, comparative learning between query and normal image prompt should incorporate both contextual and aligned residual features, rather than relying solely on residual features. AdaptCLIP treats CLIP models as a foundational service, adding only three simple adapters, visual adapter, textual adapter, and prompt-query adapter, at its input or output ends. AdaptCLIP supports zero-/few-shot generalization across domains and provides a training-free approach on target domains once trained on a base dataset. AdaptCLIP achieves state-of-the-art performance on 12 anomaly detection benchmarks from industrial and medical domains, significantly outperforming existing competitive methods.

Approximate Nearest Neighbor Search (ANNS) has become fundamental to modern deep learning applications, having gained particular prominence through its integration into recent generative models that work with increasingly complex datasets and higher vector dimensions. Existing CPU-only solutions, even the most efficient graph-based ones, struggle to meet these growing computational demands, while GPU-only solutions face memory constraints. As a solution, we propose PilotANN, a hybrid CPU-GPU system for graph-based ANNS that utilizes both CPU's abundant RAM and GPU's parallel processing capabilities. Our key innovation lies in decomposing the top-k search process into three complementary stages of increasing precision and decreasing computational cost: (i) GPU-accelerated subgraph traversal using SVD-reduced vectors; (ii) CPU refinement; and (iii) precise search using complete vectors. Furthermore, we introduce fast entry selection to improve search starting points while maximizing GPU utilization. Experimental results demonstrate that PilotANN achieves 3.9 -- 5.4× speedup in throughput on 100-million scale datasets, and is able to handle datasets up to 12 × larger than the GPU memory.

Flaky tests refer to test cases that exhibit inconsistent behaviors across multiple executions, potentially passing or failing unpredictably. They are frequently associated with suboptimal design practices that testers may utilize when crafting test cases, which undermine the quality of software testing. So, identifying the root causes of flaky tests is crucial for fixing them. Currently, inspired by the success of the Large Language Models (LLMs), researchers leverage the pre-trained language model to embed flaky test code as vectors and predict its root cause category based on vector similarity measures. However, such code embeddings generated by LLM mainly focus on capturing general semantic features but lack sufficient comprehension of the behavioral patterns involved in test scenarios, resulting in poor root cause identification. Test smells, which reflect poor coding practices or habits when writing test cases, provide complementary information in the root cause identification of test flakiness. Therefore, this paper proposes a root cause identification method for flaky tests based on test smells. Test smells are used to abstract and express behavioral patterns of test codes, and general semantic features extracted by vector embeddings to enhance the feature representation of flaky tests. Furthermore, to capture the complex nonlinear relationships between test smell features and code embeddings, a Feedforward Neural Network is constructed to categorize the root cause of test flakiness. To validate the effectiveness of our method, we performed evaluations on a dataset consisting of 451 Java flaky test cases. The experimental results indicate that our method achieves an F1-score of 80%, which is 7% higher than that of the baseline model that does not incorporate test smells.

Flaky tests can non-deterministically pass or fail on the same version of code. Their occurrence prevents using test results to determine if there are bugs in the program, thereby reducing the credibility of software testing. Concurrency is a main cause of flaky tests, where different concurrent thread interleaving may lead to uncertain program execution results. Essentially, inconsistent test execution results are more likely to occur when threads with data dependencies are interleaved. But the existing work scheduled threads randomly with no guide during flaky test detection, resulting in a vast interleaving space and low efficiency. To address this issue, we propose a data conflict-guided flaky test detection approach, which prioritizes threads with data dependency for scheduling to try to alter the test results, thereby detecting flaky tests more effectively. In more detail, the dynamic execution trace of test cases is tracked in multithreaded programs. Then, by analyzing read-write operations in the trace, data conflicts are identified. On this basis, a Bayesian Network is put forward to evaluate the degree of impact of data conflicts on test flakiness, which implies potential flaky risk. That is, the greater the flaky risk, the more likely the data conflict is to cause a flaky test. So, the scheduling strategy prioritizes threads with data conflict pairs that have a significant impact on test flakiness, to trigger test flakiness as soon as possible. To verify the effectiveness of our approach, we conduct experiments on multithreaded Java programs. The experimental results show that our approach can discover the flakiness of tests within an average of 118 seconds, and the accuracy in detecting flaky tests can reach 75 %.

Community detection (CD) is a classic graph inference task that partitions nodes of a graph into densely connected groups. While many CD methods have been proposed with either impressive quality or efficiency, balancing the two aspects remains a challenge. This study explores the potential of deep graph learning to achieve a better trade-off between the quality and efficiency of K-agnostic CD, where the number of communities K is unknown. We propose PRoCD (Pre-training & Refinement fOr Community Detection), a simple yet effective method that reformulates K-agnostic CD as the binary node pair classification. PRoCD follows a pre-training & refinement paradigm inspired by recent advances in pre-training techniques. We first conduct the offline pre-training of PRoCD on small synthetic graphs covering various topology properties. Based on the inductive inference across graphs, we then generalize the pre-trained model (with frozen parameters) to large real graphs and use the derived CD results as the initialization of an existing efficient CD method (e.g., InfoMap) to further refine the quality of CD results. In addition to benefiting from the transfer ability regarding quality, the online generalization and refinement can also help achieve high inference efficiency, since there is no time-consuming model optimization. Experiments on public datasets with various scales demonstrate that PRoCD can ensure higher efficiency in K-agnostic CD without significant quality degradation.

Tests may pass or fail under the same conditions. These tests are commonly known as flaky tests. In Android applications, the primary reason for flaky tests is attributed to its event-driven programming paradigm and multi-threading concurrency mechanism. It may activate an unexpected event order when a test is executed, causing test flakiness. The later the execution of asynchronous events, the more likely it is to result in test flakiness. Inspired by this deduction, this paper puts forward a flaky test detection method for concurrent Android applications based on adaptive latest position execution. In more detail, the latest execution positions of each asynchronous event are identified by analyzing the sequential dependencies between events. On this basis, the asynchronous event is scheduled at the corresponding position, thereby trying to change the test results and detecting flaky tests. To validate the effectiveness and efficiency of our approach, a series of experiments are conducted on 16 known flaky test cases across 7 Android applications. The experimental results show that compared with the state-of-the-art tool FlakeScanner, the flaky test detection rate of our approach improves by 18.75%.

Graph partitioning is crucial in distributed graph-parallel computing systems, and it is challenging for graph partitioning to optimize the communication cost and load balancing together. Existing state-of-the-art works, such as Powerlyra and TopoX, optimize the load balancing by randomly distributing the edges of high-degree vertices, which inevitably brings a high communication cost that is unbounded. This paper proposes a graph partition model that can minimize communication cost while maximizing load balancing. More specifically, we model the graph partition as the combinatorial design problem. Our proposed model can provide high-quality partition that guarantees that the computing load can be evenly distributed to each worker and minimizes the communication cost with a near-optimal theoretical boundary.Based on the proposed model, we extend the hybrid-cut partitioning algorithm for the power-law graph and propose HCPD, a hybrid-cut partitioning algorithm based on combinatorial design. HCPD uses the proposed model to optimize the load balancing and communication cost simultaneously for high-degree vertices, and assigns the high-degree vertices and their low-degree neighbors to the same workers by label propagation to reduce the overall communication cost. In this way, we partition the low-degree and high-degree vertices holistically and further improve the partition quality, unlike Powerlyra and TopoX, which deal with the two parts independently. Our experiments show that HCPD outperforms Powerlyra on PageRank task by up to 2× faster on real-world power-law graphs with billions of edges.

Graph partitioning is crucial in distributed graph computing systems, while impacting load balancing and communication between machines. To cope with the soaring scale of graphs, the streaming model has shown promising performance in graph partitioning. Although streaming model can deal with the bottleneck of memory usage for large-scale graphs, existing streaming partitioning algorithms not only lack sufficient quality but also cannot provide theoretical boundaries for graph partitioning. In addition, most streaming partitioning algorithms are sensitive to the order of edge streaming. In this paper, we model the edge partitioning problem as a combinatorial design problem, and provide a tight theoretical boundary. Based on the balanced edge partitioning design, we proposed a mixed-state streaming edge partitioning algorithm, which can generate high-quality graph partitions by mapping matrix and use the historical partition information to further optimize the partition quality and load balance. The experiments show that our proposed algorithm reduces partitioning time by more than half compared to the mainstream HDRF algorithm while maintaining load balance, and improves partitioning quality by about three times.

SimRank is an important metric to measure the topological similarity between two nodes in a graph. In particular, single-source and top-k SimRank has numerous applications in recommendation systems, network analysis, and web mining, etc. Mathematically, given a vertex, the computation of single-machine and single-source SimRank mainly lies in matrix-matrix operations. However, it is almost impossible to directly compute on large graphs. Thus, existing works yield to two main operations: a series of random walks, and sparse matrix and dense vector multiplication operations. This brings about high computation cost for SimRank on large graphs. In real-world applications, there is always the query time and accuracy trade-off, which hinders the computation of high-precision SimRank on large-scale graphs. To handle this problem, this paper proposesClipSim, the first GPU-friendly parallel framework that accelerates the single-source SimRank on GPU with accuracy guarantee. We design a novel data structure and GPU-friendly parallel algorithms for efficient computation of all the operations of SimRank on GPU. Moreover, our theoretical derivation enables ClipSim to largely reduce the number of random walks required for each node, while maintaining the same theoretical accuracy as the state-of-the-art algorithm, ExactSim. We conduct extensive experiments on real-world and synthetic datasets to demonstrate the accuracy and efficiency of ClipSim. The results show that compared with ExactSim, ClipSim obtains single-source SimRank vectors with the same accuracy and up to 160× faster computation time.

The goal of few-shot fine-grained image classification is to recognize rarely seen fine-grained objects in the query set, given only a few samples of this class in the support set. Previous works focus on learning discriminative image features from a limited number of training samples for distinguishing various fine-grained classes, but ignore one important fact that spatial alignment of the discriminative semantic features between the query image with arbitrary changes and the support image, is also critical for computing the semantic similarity between each support-query pair. In this work, we propose an object-aware long-short-range spatial alignment approach, which is composed of a foreground object feature enhancement (FOE) module, a long-range semantic correspondence (LSC) module and a short-range spatial manipulation (SSM) module. The FOE is developed to weaken background disturbance and encourage higher foreground object response. To address the problem of long-range object feature misalignment between support-query image pairs, the LSC is proposed to learn the transferable long-range semantic correspondence by a designed feature similarity metric. Further, the SSM module is developed to refine the transformed support feature after the long-range step to align short-range misaligned features (or local details) with the query features. Extensive experiments have been conducted on four benchmark datasets, and the results show superior performance over most state-of-the-art methods under both 1-shot and 5-shot classification scenarios.

ABSTRACT: In plants, 24 nucleotide small interfering RNAs (24-nt siRNAs) account for a large percentage of the total siRNA pool, and they play an important role in guiding plant-specific RNA-directed DNA methylation (RdDM), which transcriptionally silences transposon elements, transgenes, repetitive sequences and some endogenous genes. Several loci in plant genomes produce clusters of 24-nt RNAs, and these loci are receiving increasing attention from the research community. However, at present there is no bioinformatics resource dedicated to 24-nt siRNA loci and their derived 24-nt siRNAs. Thus, in this study, Pln24NT, a freely available web resource, was created to centralize 24-nt siRNA loci and 24-nt siRNA information, including fundamental locus information, expression profiles and annotation of transposon elements, from next-generation sequencing (NGS) data for 10 popular plant species. An intuitive web interface was also developed for convenient searching and browsing, and analytical tools were included to help users flexibly analyze their own siRNA NGS data. Pln24NT will help the plant research community to discover and characterize 24-nt siRNAs, and may prove useful for studying the roles of siRNA in RNA-directed DNA methylation in plants. AVAILABILITY AND IMPLEMENTATION: http://bioinformatics.caf.ac.cn/Pln24NT . CONTACT: [email protected]. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.