Institution profile

Hanyang University

Academic institutionasia · kr
Official website
Research library312linked papers
Opportunities0open roles
Selected work

Representative Papers

Contrastive Knowledge Distillation for Anomaly Detection in Multi-Illumination/Focus Display Images

Jul 23, 20232023 18th International Conference on Machine Vision and Applications (MVA)

In this paper, we tackle automatic anomaly detection in multi-illumination and multi-focus display images. The minute defects on the display surface are hard to spot out in RGB images and by a model trained with only normal data. To address this, we propose a novel contrastive learning scheme for knowledge distillation-based anomaly detection. In our framework, Multiresolution Knowledge Distillation (MKD) is adopted as a baseline, which operates by measuring feature similarities between the teacher and student networks. Based on MKD, we propose a novel contrastive learning method, namely Multiresolution Contrastive Distillation (MCD), which does not require positive/negative pairs with an anchor but operates by pulling/pushing the distance between the teacher and student features. Furthermore, we propose the blending module that transforms and aggregate multi-channel information to the three-channel input layer of MCD. Our proposed method significantly outperforms competitive state-of-the-art methods in both AUROC and accuracy metrics on the collected Multi-illumination and Multi-focus display image dataset for Anomaly Detection (MMdAD).

3 citationsRead paper

Spot-and-Scoot: Peeking Into Spot Instance Availability

Apr 08, 2026

High observation costs hinder effective monitoring of the dynamic availability of cloud spot instances. This work proposes Ding-Dong Ditch, a novel method that leverages the mechanism of immediately canceling spot instance requests upon acceptance to obtain binary availability signals at near-zero runtime cost, while estimating available capacity through concurrent requests. It is the first approach to actively probe availability using early-stage signals from cloud platform scheduling lifecycles, revealing that interruptions of the same instance type are highly synchronized within three minutes. Experiments across 68 instance types and 15 regions on AWS and Azure demonstrate that the method achieves an F1-macro score of 0.90 for current availability modeling, maintaining 0.85 even for 60-minute-ahead predictions. TPC-DS workload simulations further confirm its effectiveness in significantly reducing computational loss.

1 citationsRead paper

SweEval: Do LLMs Really Swear? A Safety Benchmark for Testing Limits for Enterprise Use

May 22, 2025North American Chapter of the Association for Computational Linguistics

Existing safety and ethics evaluations for large language models (LLMs) in enterprise multilingual, cross-cultural communication—e.g., email drafting and sales copy—lack systematic assessment of resistance to abusive instructions, cultural contextual adaptation, and ethical alignment. Method: We introduce the first enterprise-oriented, multilingual adversarial instruction safety benchmark, pioneering an explicit induction-based safety evaluation paradigm. It formalizes profanity-embedded adversarial prompts, integrates real-world communicative contexts and tonal variations, and jointly measures model refusal capability and cultural-linguistic comprehension. Our methodology includes constructing a multilingual profanity lexicon, human-in-the-loop + rule-augmented evaluation protocols, an open-source dataset, and an automated evaluation framework. Results: Evaluated on 20+ mainstream LLMs, we observe significantly degraded compliance under informal contexts. The benchmark delivers reproducible, quantitative safety metrics—enabling rigorous risk assessment and governance for enterprise AI deployment.

1 citationsRead paper

A Balanced Approach of Rapid Genetic Exploration and Surrogate Exploitation for Hyperparameter Optimization

Apr 10, 2025IEEE Access

To address the exploration-exploitation imbalance and inefficient utilization of historical evaluation data in evolutionary algorithms (EAs) for hyperparameter optimization (HPO), this paper proposes an enhanced genetic algorithm (GA) framework integrated with a lightweight linear surrogate model. The linear surrogate is seamlessly embedded into GA’s selection and mutation operators, enabling a population-driven hybrid search strategy and performance-feedback-driven adaptive model updating—without requiring gradient computation or expensive modeling overhead. This design dynamically balances global exploration and local exploitation. On standard HPO benchmarks, the method achieves an average performance improvement of 1.89% (range: −3.45% to +6.55%) over state-of-the-art approaches, with negligible increase in training cost. Its core contribution lies in the first structured integration of linear surrogates with genetic operations, achieving a favorable trade-off among efficiency, accuracy, and scalability.

1 citationsRead paper
Recent publications

Latest Papers