2 papers
cs.AI2026
When2Think: Learning When and How Much to Reason
Jaejun Shim, HyunJin Kim, Young Jin Kim +1
Large Reasoning Models (LRMs) often overthink easy problems and underthink hard ones, leading to inefficient computation allocation. Existing methods regulate generated computation…
cs.IR2026
Temporal Preference Optimization for Unsupervised Retrieval
HyunJin Kim, Jaejun Shim, Young Jin Kim +1
Unsupervised dense retrievers offer scalability by learning semantic similarity from unlabeled documents via contrastive learning, but they struggle to capture the temporal relevan…