Visa (United States)

Software bug prediction still represents a fundamental problem in modern software engineering as system complexity grows exponentially. This paper presents a new machine learning approach for predictive bug detection in software testing. It integrates multi-dimensional features extracted from software metric to discriminate against possible defects before release and uses advanced classification algorithms (especially deep learning algorithms) for such purposes. Statistical analysis of defects suggests the superiority of our proposed SMART (Software Metrics Analysis and Risk Tracking) model in terms of the accuracy of prediction of defects, as with synthetic datasets involving 5,000 cases with twelve-dimensional features. The experimental results demonstrate marked improvement in fault detection rate while preserving computational efficiency. Filters applied within the framework implementation can compensate for existing drawbacks of real-time defect prediction approaches, thus providing practical advice to quality assurance processes. This paper serves as a building block to support automated intelligent software testing and improved quality assurance.

The financial industry leverages machine learning in more ways than just finding the right alpha signal. It grapples with supply chains, business processes, marketing, churn, fraud, and money laundering, all while maintaining compliance with the various regulatory frameworks it is beholden to. Due to the sheer volume of wealth being handled by the financial industry and its critical role in everyday life, it has been a lucrative target for a wide spectrum of ever-evolving bad actors. With each successive iteration of this workshop, we have attempted to capture the breadth of these actors - fraudsters, money launderers, market manipulators, and potentially nation-state-level risks. The emerging advances in Generative AI make this a particularly exciting time to host this workshop. GenAI offers groundbreaking approaches to handling the various data types prevalent in the financial sector. From a security point of view, bad actors are actively using Generative AI creatively to thwart conventional defenses (e.g. voice cloning, better synthetic identities), and this workshop's audience would benefit from commonly applicable defenses & best practices against such threats. Last but not the least, there is now an increasing willingness from the financial industry towards deeper engagement and data sharing with academia.

We present unconditionally perfectly secure protocols in the semi-honest setting for several functionalities: (1) private elementwise equality; (2) private bitwise integer comparison; and (3) bit-decomposition. These protocols are built upon a new concept called Shared Oblivious Transfer (Shared OT). Shared OT extends the one-out-of-N String OT by replacing strings with integers modulo M and allowing additive secret-sharing of all inputs and outputs. These extensions can be implemented by simple local computations without incurring additional OT invocations. We believe our Shared OT may be of independent interest. Our protocols demonstrate the best round, communication, and computational complexities compared to all other protocols secure in a similar setting. Moreover, all of our protocols involve either 2 or 3 rounds.

Bipartite graphs are ubiquitous in real-world applications, where butterflies (2x2-bicliques) serve as fundamental building blocks for analyzing graph motifs. While existing wedge aggregation based butterfly computation methods are effective for small-scale graphs, they face significant scalability challenges when processing large-scale bipartite graphs with billions of vertices and edges. In this paper, we present MONARCH, a distributed framework for efficient butterfly computation. Unlike traditional approaches that require expensive two-hop neighbor traversal, MONARCH operates exclusively on first-hop neighbor information, reducing both computation time and communication overhead. This pioneering framework expands the boundary of butterfly computation methods, making it feasible to process bipartite graphs that are orders of magnitude larger. To demonstrate its scalability, we show experimental results on bipartite graph with billions of vertices and edges.

Traffic forecasting, crucial for urban planning, requires accurate predictions of spatial-temporal traffic patterns across urban areas. Existing research mainly focuses on designing complex spatial-temporal models to capture these dependencies. However, this field faces challenges related to data scarcity and model stability, which results in limited performance improvement. To address these issues, we propose Spatial-Temporal Masked AutoEncoders (STMAE), a plug-and-play framework designed to enhance existing spatial-temporal models on traffic prediction. STMAE operates in two stages. In the pretraining stage, an encoder processes partially visible traffic data produced by a dual-masking strategy, including biased random walk-based spatial masking and patch-based temporal masking. Subsequently, two decoders aim to reconstruct the masked counterparts from both spatial and temporal perspectives. The fine-tuning stage retains the pretrained encoder and integrates it with decoders from existing backbones to improve traffic forecasting accuracy. Our results on traffic benchmarks show that STMAE can largely enhance the forecasting capabilities of various spatial-temporal models.

In modern data-intensive enterprises, ensuring reliability, accuracy, and trust in data pipelines is critical for downstream analytics and decision-making. Traditional monitoring systems often struggle to detect complex dependencies and subtle anomalies across distributed data ecosystems, leading to delayed remediation and increased operational risk. This research introduces a hybrid AI-driven data observability framework that combines Graph Neural Networks (GNNs) for structural dependency modeling with Bayesian anomaly detection for probabilistic uncertainty estimation. The proposed approach leverages GNNs to capture high-dimensional relational patterns across datasets, workflows, and metadata, while Bayesian inference provides robust anomaly detection under uncertainty and sparse conditions. Experiments conducted on enterprise-scale synthetic and real-world datasets demonstrate significant improvements in anomaly detection accuracy, early incident identification, and false-positive reduction compared to conventional rule-based and statistical monitoring techniques. By unifying structural learning with probabilistic reasoning, this hybrid framework enhances trust in data pipelines, reduces mean-time-to-detection (MTTD), and ensures resilient data operations. The findings highlight the potential of AI-driven observability as a foundational capability for next-generation data reliability engineering.

Recent advancements in pre-trained language-image models have ushered in a new era of visual comprehension. Leveraging the power of these models, this article tackles two issues within the realm of visual analytics: (1) the efficient exploration of large-scale image datasets and identification of data biases within them; (2) the evaluation of image captions and steering of their generation process. On the one hand, by visually examining the captions generated from language-image models for an image dataset, we gain deeper insights into the visual contents, unearthing data biases that may be entrenched within the dataset. On the other hand, by depicting the association between visual features and textual captions, we expose the weaknesses of pre-trained language-image models in their captioning capability and propose an interactive interface to steer caption generation. The two parts have been coalesced into a coordinated visual analytics system, fostering the mutual enrichment of visual and textual contents. We validate the effectiveness of the system with domain practitioners through concrete case studies with large-scale image datasets.