The 9th Social Media Prediction Challenge

News

• [2026-03-24] Please remember to register your team with full information here: SMP-Image or SMP-Video.
• [2026-03-20] The challenge website is opened.
• [2026-03-10] We will host the 9th Social Media Prediction Challenge with two tracks this year.

Overview

SMP Challenge is an annual challenge that seeks excellent research teams or solutions for social multimodal prediction and meaningfully improving people’s social lives and business scenarios. The challenges for researchers looking at their models on social media data are large-scale, multimodal, and multivariate.

Social Media Prediction (SMP). With the ever-changing public attention and individual interests, predicting the exact values accurately of online popularity is even more important than before. Social Media Popularity Prediction focuses on predicting the impact of sharing different posts for a publisher on social media. This year, we will have two tracks SMP-Image and SMP-Video for social images and videos.

Benchmark

Social Media Prediction Dataset (SMPD) consists of SMPD-Image and SMPD-Video. The former is a large-scale social image dataset, containing over 680K social images and 80K users. The latter is a large-scale social short-form video dataset with 6K videos and 4.5K users.

Important Dates

Citation

If you like or use the resources from the challenge, please read and cite the following references:

@inproceedings{SMP2023,
  title={SMP Challenge: An Overview and Analysis of Social Media Prediction Challenge},
  author={Wu, Bo and Liu, Peiye and Cheng, Wen-Huang and Liu, Bei and Zeng, Zhaoyang and Wang, Jia and Huang, Qiushi and Luo, Jiebo},
  booktitle={Proceedings of the 31th ACM International Conference on Multimedia},
  year={2023}}
@inproceedings{Wu2017DTCN,
    title={Sequential Prediction of Social Media Popularity with Deep Temporal Context Networks},
    author={Wu, Bo and Cheng, Wen-Huang and Zhang, Yongdong and Qiushi, Huang and Jintao, Li and Mei, Tao},
    booktitle={International Joint Conference on Artificial Intelligence (IJCAI)},
    year={2017}}
@inproceedings{Wu2016TemporalPrediction,
    author = {Wu, Bo and Mei, Tao and Cheng, Wen-Huang and Zhang, Yongdong},
    title = {Unfolding Temporal Dynamics: Predicting Social Media Popularity Using Multi-scale Temporal Decomposition},
    booktitle = {Proceedings of the Thirtieth AAAI Conference on Artificial Intelligence (AAAI)}
    year = {2016}}
@inproceedings{wu2016time,
  title={Time matters: Multi-scale temporalization of social media popularity},
  author={Wu, Bo and Cheng, Wen-Huang and Zhang, Yongdong and Mei, Tao},
  booktitle={Proceedings of the 24th ACM international conference on Multimedia},
  year={2016}}
            

Our Team

Bo Wu

MIT-IBM Watson AI Lab

Peiye Liu

ByteDance

Zhaoyang Zeng

International Digital Economy Academy

Jia Wang

National Yang Ming Chiao Tung University

Qiushi Huang

University of Surrey

Bei Liu

Microsoft Research Asia

Jiebo Luo

University of Rochester

Wen-Huang Cheng

National Taiwan University