Sample data and trained model weights for M-BridgeNet, a framework for detecting
cross-platform bridge posts — posts that carry a narrative from one Chinese
social platform to another (Weibo, Douyin, Bilibili, Zhihu).
⚠️ This is a truncated SAMPLE, not the full dataset. Each event JSON is capped
to ≤200 posts / candidate pairs to protect the full data asset. The sample
illustrates the format and supports light experimentation, but it… See the full description on the dataset page:
https://huggingface.co/datasets/weifar/M-BridgeNet-ch.