Large language model (LLM) agents are moving beyond reactive responses toward proactive assistance that anticipates user needs without explicit requests. How…
机构:Columbia
来源:arXiv 2610.27400 | AI4Papers 论文推荐平台