Feng Mei
Research Subjects
Research Interests
Feng Mei has a total of 13 co-authors. Here are the top co-authors by number of shared publications:
- Sulong Xu
JD.com, Beijing, China
3 shared publications - Xiwei Zhao
JD.com, Beijing, China
3 shared publications - Dongyue Wang
JingDong Group, Beijing, China
2 shared publications - Jinghe Hu
JD.com, Beijing, China
2 shared publications - Binglei Zhao
JD.com, Beijing, China
1 shared publication - Guang Xu
JD.com, Inc, Beijing, China
1 shared publication
Latest Publications
- Open Access
- 13 July 2025
- Published by Association for Computing Machinery (ACM)
- p. 2880-2884
Hierarchical User Long-term Behavior Modeling for Click-Through Rate Prediction
- 13 July 2025
- Published by Association for Computing Machinery (ACM)
- p. 2880-2884
Open AccessHierarchical User Long-term Behavior Modeling for Click-Through Rate Prediction
- Open Access
- 8 May 2025
- Published by Association for Computing Machinery (ACM)
- p. 621-630
A Hybrid Cross-Stage Coordination Pre-ranking Model for Online Recommendation Systems
- 8 May 2025
- Published by Association for Computing Machinery (ACM)
- p. 621-630
Open AccessA Hybrid Cross-Stage Coordination Pre-ranking Model for Online Recommendation Systems
- Open Access
- 13 May 2024
- Published by Association for Computing Machinery (ACM)
- p. 311-318
PPM : A Pre-trained Plug-in Model for Click-through Rate Prediction
Click-through rate (CTR) prediction is a core task in recommender systems. Existing methods (IDRec for short) rely on unique identities to represent distinct users and items that have prevailed for decades. On one hand, IDRec often faces significant performance degradation on cold-start problem; on the other hand, IDRec cannot use longer training data due to constraints imposed by iteration efficiency. Most prior studies alleviate the above problems by introducing pre-trained knowledge(e.g. pre-trained user model or multi-modal embeddings). However, the explosive growth of online latency can be attributed to the huge parameters in the pre-trained model. Therefore, most of them cannot employ the unified model of end-to-end training with IDRec in industrial recommender systems, thus limiting the potential of the pre-trained model. To this end, we propose a pre-trained plug-in CTR model, namely PPM. PPM employs multi-modal features as input and utilizes large-scale data for pre-training. Then, PPM is plugged in IDRec model to enhance unified model's performance and iteration efficiency. Upon incorporating IDRec model, certain intermediate results within the network are cached, with only a subset of the parameters participating in training and serving. Hence, our approach can successfully deploy an end-to-end model without causing huge latency increases. Comprehensive offline experiments and online A/B testing at JD E-commerce demonstrate the efficiency and effectiveness of PPM.
- 13 May 2024
- Published by Association for Computing Machinery (ACM)
- p. 311-318
Open AccessPPM : A Pre-trained Plug-in Model for Click-through Rate Prediction
Click-through rate (CTR) prediction is a core task in recommender systems. Existing methods (IDRec for short) rely on unique identities to represent distinct users and items that have prevailed for decades. On one hand, IDRec often faces significant performance degradation on cold-start problem; on the other hand, IDRec cannot use longer training data due to constraints imposed by iteration efficiency. Most prior studies alleviate the above problems by introducing pre-trained knowledge(e.g. pre-trained user model or multi-modal embeddings). However, the explosive growth of online latency can be attributed to the huge parameters in the pre-trained model. Therefore, most of them cannot employ the unified model of end-to-end training with IDRec in industrial recommender systems, thus limiting the potential of the pre-trained model. To this end, we propose a pre-trained plug-in CTR model, namely PPM. PPM employs multi-modal features as input and utilizes large-scale data for pre-training. Then, PPM is plugged in IDRec model to enhance unified model's performance and iteration efficiency. Upon incorporating IDRec model, certain intermediate results within the network are cached, with only a subset of the parameters participating in training and serving. Hence, our approach can successfully deploy an end-to-end model without causing huge latency increases. Comprehensive offline experiments and online A/B testing at JD E-commerce demonstrate the efficiency and effectiveness of PPM.