Generative Representational Learning of Foundation Models for Recommendation
Abstract
Developing a single foundation model with the capability to excel across diverse tasks has been a long-standing objective in the field of artificial intelligence. As the wave of general-purpose foundation models sweeps across various domains, their influence has significantly extended to the field of recommendation systems. While recent efforts have explored recommendation foundation models for various generative tasks, they often overlook crucial embedding tasks and struggle with the complexities of multi-task learning, including knowledge sharing & conflict resolution, and convergence speed inconsistencies. To address these limitations, we introduce RecFound, a generative representational learning framework for recommendation foundation models. We construct the first comprehensive dataset for recommendation foundation models covering both generative and embedding tasks across diverse scenarios. Based on this dataset, we propose a novel multi-task training scheme featuring a Task-wise Mixture of Low-rank Experts (TMoLE) to handle knowledge sharing & conflict, a Step-wise Convergence-oriented Sample Scheduler (S2Sched) to address inconsistent convergence, and a Model Merge module to balance the performance across tasks. Experiments demonstrate that RecFound achieves state-of-the-art performance across various recommendation tasks, outperforming existing baselines.
Cite
@article{arxiv.2506.11999,
title = {Generative Representational Learning of Foundation Models for Recommendation},
author = {Zheli Zhou and Chenxu Zhu and Jianghao Lin and Bo Chen and Ruiming Tang and Weinan Zhang and Yong Yu},
journal= {arXiv preprint arXiv:2506.11999},
year = {2025}
}
Comments
Project page is available at https://junkfood436.github.io/RecFound/