1 paper · 1 filter
Jingmin An, Wei Liu, Qian Wang +1
Time functions as a fundamental dimension of human cognition, yet the mechanisms by which Large Language Models (LLMs) encode chronological progression remain opaque. We demonstrat…