OpenOneRec Technical Report

作者：Guorui Zhou, Honghui Bao, Jiaming Huang, Jiaxin Deng, Jinghao Zhang, Junda She, Kuo Cai, Lejian Ren, Lu Ren, Qiang Luo, Qianqian Wang, Qigen Hu, Rong-Qing Zhang, Ruiming Tang, Shiyao Wang, Wuchao Li, Xiangyu Wu, Xinchen Luo, Xing-Yao Wang, Yifei Hu, Yun-Jie Wu, Zhanyun Liu, Zhi-Yong Zhang, Zixin Zhang, Bo Chen, Bin Wen, Chao Ma, Cheng-bin Song, Chenglong Chu, Defu Lian, Fan Yang, Fengfeng Jiang, Hong Cheng, Huanjie Wang, Kun Gai, Pengfei Zheng, Qiang Wang, Rui Huang, Si-Yuan Mao, Tingting Gao, Wei Yuan, Yan Wang, Yangfan Zhou, Yi Su, Zexuan Cheng, Zhixin Ling, Ziming Li · 发表于：arXiv.org · 年份：2025 · DOI：10.48550/arXiv.2512.24762 · 被引用次数：10 · 研究领域：Computer Science

While the OneRec series has successfully unified the fragmented recommendation pipeline into an end-to-end generative framework, a significant gap remains between recommendation systems and general intelligence. Constrained by isolated data, they operate as domain specialists-proficient in pattern matching but lacking world knowledge, reasoning capabilities, and instruction following. This limitation is further compounded by the lack of a holistic benchmark to evaluate such integrated capabilities. To address this, our contributions are: 1) RecIF Bench&Open Data: We propose RecIF-Bench, a holistic benchmark covering 8 diverse tasks that thoroughly evaluate capabilities from fundamental prediction to complex reasoning. Concurrently, we release a massive training dataset comprising 96 million interactions from 160,000 users to facilitate reproducible research. 2) Framework&Scaling: To ensure full reproducibility, we open-source our comprehensive training pipeline, encompassing data processing, co-pretraining, and post-training. Leveraging this framework, we demonstrate that recommendation capabilities can scale predictably while mitigating catastrophic forgetting of general knowledge. 3) OneRec-Foundation: We release OneRec Foundation (1.7B and 8B), a family of models establishing new state-of-the-art (SOTA) results across all tasks in RecIF-Bench. Furthermore, when transferred to the Amazon benchmark, our models surpass the strongest baselines with an average 26.8% improvement...