Hyunsuk Chung — accepted AI-conference papers

2026

MulTiCast: A Multimodal Time Series Forecasting System

AAAI 2026technical

Time-series forecasting plays an essential role in domains such as finance, healthcare, and energy. Yet most existing systems operate in a unimodal setting, overlooking complementary information available from visual and textual modalities. There are surprisingly few time-series forecasting demos, a

Cited by 0 Source ↗PDF ↗Scholar ↗

2024

PEACH: Pretrained-Embedding Explanation across Contextual and Hierarchical Structure

Feiqi Cao, Soyeon Caren Han, Hyunsuk Chung

IJCAI 2024poster

In this work, we propose a novel tree-based explanation technique, PEACH (Pretrained-embedding Explanation Across Contextual and Hierarchical Structure), that can explain how text-based documents are classified by using any pretrained contextual embeddings in a tree-based human-interpretable manner.…

Cited by 0 Source ↗PDF ↗code ↗Scholar ↗