Large language models (LLMs) have demonstrated strong reasoning capabilities, motivating their use as the backbone of foundation recommendation models (FRMs)…
机构:快手
来源:arXiv 2610.30675 | AI4Papers 论文推荐平台