JinYoung Lee

Youngmoon Jung, Jinyoung Lee, Seungjin Lee et al.

Recent advances in flexible keyword spotting (KWS) with text enrollment allow users to personalize keywords without uttering them during enrollment. However, there is still room for improvement in target keyword performance. In this work, we propose a novel few-shot transfer learning method, called text-aware adapter (TA-adapter), designed to enhance a pre-trained flexible KWS model for specific keywords with limited speech samples. To adapt the acoustic encoder, we leverage a jointly pre-trained text encoder to generate a text embedding that acts as a representative vector for the keyword. By fine-tuning only a small portion of the network while keeping the core components' weights intact, the TA-adapter proves highly efficient for few-shot KWS, enabling a seamless return to the original pre-trained model. In our experiments, the TA-adapter demonstrated significant performance improvements across 35 distinct keywords from the Google Speech Commands V2 dataset, with only a 0.14% increase in the total number of parameters.

Seoik Jung, Taekyung Song, Joshua Jordan Daniel et al.

Video Anomaly Detection (VAD) is critical for surveillance and public safety. However, existing benchmarks are limited to either frame-level or video-level tasks, restricting a holistic view of model generalization. This work first introduces a softmax-based frame allocation strategy that prioritizes anomaly-dense segments while maintaining full-video coverage, enabling balanced sampling across temporal scales. Building on this process, we construct two complementary benchmarks. The image-based benchmark evaluates frame-level reasoning with representative frames, while the video-based benchmark extends to temporally localized segments and incorporates an abnormality scoring task. Experiments on UCF-Crime demonstrate improvements at both the frame and video levels, and ablation studies confirm clear advantages of anomaly-focused sampling over uniform and random baselines.

Ali Tamimi, Ozgur Oksuz, Jinyoung Lee et al.

Cybersecurity risks are often managed by reducing the system's attack surface, which includes minimizing the number of interconnections, privileges, and impacts of an attack. While attack surface reduction techniques have been frequently deployed in more traditional information technology (IT) domains, metrics tailored to cyber-physical systems (CPS) have not yet been identified. This paper introduces attack surface analysis metrics and algorithms to evaluate the attack surface of a CPS. The proposed approach includes both physical system impact metrics, along with a variety of cyber system properties from the software (network connections, methods) and operating system (privileges, exploit mitigations). The proposed algorithm is defined to incorporate with the Architecture Analysis \& Design Language (AADL), which is commonly used to many CPS industries to model their control system architecture, and tools have been developed to automate this analysis on an AADL model. Furthermore, the proposed approach is evaluated on a distribution power grid case study, which includes a 7 feeder distribution system, AADL model of the SCADA control centers, and analysis of the OpenDNP3 protocol library used in many real-world SCADA systems.