2 papers
cs.LG2026
Counting on Thinking: Tracing Evidence Integration in Language Models
Jingming Xue, Robert C. Wilson, Huadong Xiong
Finite computational resources force a tradeoff between automatic System 1 processes and costly System 2 thinking. Large language models (LLMs) can spend extra computation on hard…
cs.LG2026
Thinking Under Uncertainty: Evidence Use and Information-Seeking in Language Models
Hua-Dong Xiong, Xinyuan Yan, Ji-An Li +3
Inference-time thinking improves the performance of large language models, but aggregate outcomes do not reveal whether models use available evidence more effectively or seek infor…