Institution profile

Cheju National University

Academic institutionasia · kr
Official website
Research library9linked papers
Opportunities0open roles
Selected work

Representative Papers

Risk-Constrained Freshness-Aware Semantic Caching for Open-Web Retrieval-Augmented LLMs

Jul 05, 2026

This work addresses the critical limitation of existing semantic caching approaches, which neglect the temporal dynamics of evidence freshness in open-web settings, often yielding outdated results. To overcome this, we propose FreshCache, the first framework to formulate cache reuse as a risk-constrained sequential inference problem. FreshCache integrates an exponential decay model with a multilayer perceptron (MLP) to predict the probability of cache entry staleness and employs a three-tier risk-gating mechanism for fine-grained freshness control, enabling cache entries to degrade gracefully over time rather than fail in a binary manner. We introduce FreshCache-Bench, a novel benchmark based on real-world web snapshots, to evaluate freshness-aware caching. Experiments demonstrate that within a 24-hour window, FreshCache_MLP reduces search API calls by 97% while maintaining a hash-level staleness error rate of merely 0.1%, with only 0.034% of errors materially affecting answer correctness—substantially outperforming current methods.

0 citationsRead paper

Sensory-Aware Sequential Recommendation via Review-Distilled Representations

Mar 03, 2026

This work addresses the limitation of conventional sequential recommendation models in capturing users’ sensory experiences. To bridge this gap, the authors propose ASEGR, a novel framework that uniquely integrates large language models with knowledge distillation to extract structured sensory attributes—such as color and scent—from product reviews. These attributes are distilled into fixed-dimensional sensory embeddings and seamlessly incorporated into mainstream sequential recommenders like SASRec and BERT4Rec. Extensive experiments on four Amazon datasets demonstrate that ASEGR significantly outperforms baseline methods in recommendation performance while generating interpretable sensory attributes aligned with human perception, thereby validating the efficacy of sensory semantics as a complementary signal for user behavior modeling.

0 citationsRead paper

An Unsupervised Tensor-Based Domain Alignment

Jan 26, 2026

This work addresses the challenge of distributional discrepancy between source and target domains in unsupervised tensor domain adaptation by proposing a novel method that jointly optimizes alignment matrices and a shared invariant subspace. By imposing constraints on the more flexible oblique manifold—rather than the conventional Stiefel manifold—and incorporating a variance-preserving regularizer to enhance robustness, the proposed framework generalizes existing tensor alignment approaches while significantly improving both domain adaptation efficiency and classification accuracy. Extensive experiments demonstrate that the method consistently outperforms state-of-the-art techniques across multiple benchmark datasets.

0 citationsRead paper
Recent publications

Latest Papers

Risk-Constrained Freshness-Aware Semantic Caching for Open-Web Retrieval-Augmented LLMs

Jul 05, 2026

This work addresses the critical limitation of existing semantic caching approaches, which neglect the temporal dynamics of evidence freshness in open-web settings, often yielding outdated results. To overcome this, we propose FreshCache, the first framework to formulate cache reuse as a risk-constrained sequential inference problem. FreshCache integrates an exponential decay model with a multilayer perceptron (MLP) to predict the probability of cache entry staleness and employs a three-tier risk-gating mechanism for fine-grained freshness control, enabling cache entries to degrade gracefully over time rather than fail in a binary manner. We introduce FreshCache-Bench, a novel benchmark based on real-world web snapshots, to evaluate freshness-aware caching. Experiments demonstrate that within a 24-hour window, FreshCache_MLP reduces search API calls by 97% while maintaining a hash-level staleness error rate of merely 0.1%, with only 0.034% of errors materially affecting answer correctness—substantially outperforming current methods.

0 citationsRead paper

Sensory-Aware Sequential Recommendation via Review-Distilled Representations

Mar 03, 2026

This work addresses the limitation of conventional sequential recommendation models in capturing users’ sensory experiences. To bridge this gap, the authors propose ASEGR, a novel framework that uniquely integrates large language models with knowledge distillation to extract structured sensory attributes—such as color and scent—from product reviews. These attributes are distilled into fixed-dimensional sensory embeddings and seamlessly incorporated into mainstream sequential recommenders like SASRec and BERT4Rec. Extensive experiments on four Amazon datasets demonstrate that ASEGR significantly outperforms baseline methods in recommendation performance while generating interpretable sensory attributes aligned with human perception, thereby validating the efficacy of sensory semantics as a complementary signal for user behavior modeling.

0 citationsRead paper

An Unsupervised Tensor-Based Domain Alignment

Jan 26, 2026

This work addresses the challenge of distributional discrepancy between source and target domains in unsupervised tensor domain adaptation by proposing a novel method that jointly optimizes alignment matrices and a shared invariant subspace. By imposing constraints on the more flexible oblique manifold—rather than the conventional Stiefel manifold—and incorporating a variance-preserving regularizer to enhance robustness, the proposed framework generalizes existing tensor alignment approaches while significantly improving both domain adaptation efficiency and classification accuracy. Extensive experiments demonstrate that the method consistently outperforms state-of-the-art techniques across multiple benchmark datasets.

0 citationsRead paper