Preprints

Uncertainty-Aware Variational Reward Factorization via Probabilistic Preference Bases for LLM Personalization [link]

Gyuseok Lee, Wonbin Kweon, Zhenrui Yue, SeongKu Kang, Jiawei Han, Dong Wang

• We introduce Variational Reward Factorization (VRF), an uncertainty-aware framework that represents user preference as a variational distribution.

MultiCube-RAG for Multi-hop Question Answering [link]

Jimeng Shi, Wei Hu, Runchu Tian, Bowen Jin, Wonbin Kweon, SeongKu Kang, Yunfan Kang, Dingqi Ye, Sizhe Zhou, Shaowen Wang, Jiawei Han

• We propose MultiCube-RAG, a training-free method consisting of multiple cubes for multi-step reasoning and retrieval.

Improving Scientific Document Retrieval with Academic Concept Index [link]

Jeyun Lee*, Junhyoung Lee*, Wonbin Kweon, Bowen Jin, Yu Zhang, Susik Yoon, Dongha Lee, Hwanjo Yu, Jiawei Han, SeongKu Kang

• We introduce an academic concept index that represents structured knowledge of a scientific corpus, and propose a query simulation and context extension method based on this index. 

LLM-Based Compact Reranking with Document Features for Scientific Retrieval [link]

Runchu Tian, Xueqiang Xu, Bowen Jin, SeongKu Kang, Jiawei Han

• We propose CORANK, a training-free, model-agnostic reranking framework for scientific retrieval.

Graph Signal Processing for Cross-Domain Recommendation [link]

Jeongeun Lee, SeongKu Kang, Won-Yong Shin, Jeongwhan Choi, Noseong Park, Dongha Lee

• We propose CGSP, a unified cross-domain recommendation framework based on graph signal processing.