19 papers
Personalized Deep Research Query Refinement with Graph-Scaffolded Evidence Grounding
Soojin Yoon, Dongha Lee
User requests serve as research specifications for deep research agents, shaping what evidence to seek and how to synthesize it. In personalized deep research, these specifications…
Region4Web: Rethinking Observation Space Granularity for Web Agents
Donguk Kwon, Dongha Lee
Web agents perceive web pages through an observation space, yet its granularity has remained an underexamined design choice. Existing work treats observation at the same element-le…
MVIGER: Multi-View Variational Integration of Complementary Knowledge for Generative Recommender
Tongyoung Kim, Soojin Yoon, SeongKu Kang +2
Language Models (LMs) have been widely used in recommender systems to incorporate textual information of items into item IDs, leveraging their advanced language understanding and g…
Why These Documents? Explainable Generative Retrieval with Hierarchical Category Paths
Sangam Lee, Ryang Heo, SeongKu Kang +3
Generative retrieval directly decode a document identifier (i.e., docid) in response to a query, making it impossible to provide users with explanations as an answer for ``why is t…
CONDESION-BENCH: Conditional Decision-Making of Large Language Models in Compositional Action Space
Yeonjun Hwang, Sungyong Park, Minju Kim +2
Large language models have been widely explored as decision-support tools in high-stakes domains due to their contextual understanding and reasoning capabilities. However, existing…
Quantifying Genuine Awareness in Hallucination Prediction Beyond Question-Side Shortcuts
Yeongbin Seo, Dongha Lee, Jinyoung Yeo
Many works have proposed methodologies for language model (LM) hallucination detection and reported seemingly strong performance. However, we argue that the reported performance to…