Saved in:
| Main Authors: | , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2508.14948 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866909746450137088 |
|---|---|
| author | Zhang, Shangyu Quan, Shijie Wang, Zhongren Pan, Junwei Zhuang, Tianqu Fu, Bo Sun, Yilong Lin, Jieying Chen, Jushuo Li, Xiaotian Feng, Zhixiang Hu, Xian Deng, Huiting Lu, Hua Wang, Jinpeng Dai, Boqi Chen, Xiaoyu Hu, Bin Huang, Lili Wu, Yanwen Cai, Yeshou Zhou, Qi Tang, Huang Yang, Chunfeng Yin, Chengguo Jiang, Tingyu Wang, Lifeng Huang, Shudong Liu, Dapeng Xiao, Lei Gu, Haijie Xia, Shu-Tao Jiang, Jie |
| author_facet | Zhang, Shangyu Quan, Shijie Wang, Zhongren Pan, Junwei Zhuang, Tianqu Fu, Bo Sun, Yilong Lin, Jieying Chen, Jushuo Li, Xiaotian Feng, Zhixiang Hu, Xian Deng, Huiting Lu, Hua Wang, Jinpeng Dai, Boqi Chen, Xiaoyu Hu, Bin Huang, Lili Wu, Yanwen Cai, Yeshou Zhou, Qi Tang, Huang Yang, Chunfeng Yin, Chengguo Jiang, Tingyu Wang, Lifeng Huang, Shudong Liu, Dapeng Xiao, Lei Gu, Haijie Xia, Shu-Tao Jiang, Jie |
| contents | Online advertising relies on accurate recommendation models, with recent advances using pre-trained large-scale foundation models (LFMs) to capture users' general interests across multiple scenarios and tasks. However, existing methods have critical limitations: they extract and transfer only user representations (URs), ignoring valuable item representations (IRs) and user-item cross representations (CRs); and they simply use a UR as a feature in downstream applications, which fails to bridge upstream-downstream gaps and overlooks more transfer granularities. In this paper, we propose LFM4Ads, an All-Representation Multi-Granularity transfer framework for ads recommendation. It first comprehensively transfers URs, IRs, and CRs, i.e., all available representations in the pre-trained foundation model. To effectively utilize the CRs, it identifies the optimal extraction layer and aggregates them into transferable coarse-grained forms. Furthermore, we enhance the transferability via multi-granularity mechanisms: non-linear adapters for feature-level transfer, an Isomorphic Interaction Module for module-level transfer, and Standalone Retrieval for model-level transfer. LFM4Ads has been successfully deployed in Tencent's industrial-scale advertising platform, processing tens of billions of daily samples while maintaining terabyte-scale model parameters with billions of sparse embedding keys across approximately two thousand features. Since its production deployment in Q4 2024, LFM4Ads has achieved 10+ successful production launches across various advertising scenarios, including primary ones like Weixin Moments and Channels. These launches achieve an overall GMV lift of 2.45% across the entire platform, translating to estimated annual revenue increases in the hundreds of millions of dollars. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2508_14948 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Large Foundation Model for Ads Recommendation Zhang, Shangyu Quan, Shijie Wang, Zhongren Pan, Junwei Zhuang, Tianqu Fu, Bo Sun, Yilong Lin, Jieying Chen, Jushuo Li, Xiaotian Feng, Zhixiang Hu, Xian Deng, Huiting Lu, Hua Wang, Jinpeng Dai, Boqi Chen, Xiaoyu Hu, Bin Huang, Lili Wu, Yanwen Cai, Yeshou Zhou, Qi Tang, Huang Yang, Chunfeng Yin, Chengguo Jiang, Tingyu Wang, Lifeng Huang, Shudong Liu, Dapeng Xiao, Lei Gu, Haijie Xia, Shu-Tao Jiang, Jie Machine Learning Online advertising relies on accurate recommendation models, with recent advances using pre-trained large-scale foundation models (LFMs) to capture users' general interests across multiple scenarios and tasks. However, existing methods have critical limitations: they extract and transfer only user representations (URs), ignoring valuable item representations (IRs) and user-item cross representations (CRs); and they simply use a UR as a feature in downstream applications, which fails to bridge upstream-downstream gaps and overlooks more transfer granularities. In this paper, we propose LFM4Ads, an All-Representation Multi-Granularity transfer framework for ads recommendation. It first comprehensively transfers URs, IRs, and CRs, i.e., all available representations in the pre-trained foundation model. To effectively utilize the CRs, it identifies the optimal extraction layer and aggregates them into transferable coarse-grained forms. Furthermore, we enhance the transferability via multi-granularity mechanisms: non-linear adapters for feature-level transfer, an Isomorphic Interaction Module for module-level transfer, and Standalone Retrieval for model-level transfer. LFM4Ads has been successfully deployed in Tencent's industrial-scale advertising platform, processing tens of billions of daily samples while maintaining terabyte-scale model parameters with billions of sparse embedding keys across approximately two thousand features. Since its production deployment in Q4 2024, LFM4Ads has achieved 10+ successful production launches across various advertising scenarios, including primary ones like Weixin Moments and Channels. These launches achieve an overall GMV lift of 2.45% across the entire platform, translating to estimated annual revenue increases in the hundreds of millions of dollars. |
| title | Large Foundation Model for Ads Recommendation |
| topic | Machine Learning |
| url | https://arxiv.org/abs/2508.14948 |