We're making the Warpcast spam dataset public. 

Over 400,000 accounts have been processed by our model, which determines the accounts that are most likely to generate inauthentic content or unwanted notifications. 

https://github.com/warpcast/labels

That’s an exciting move, making the dataset public! Transparency around spam and inauthentic behavior detection can help researchers and developers refine their own models, identify trends, and even collaborate to tackle these challenges more effectively.

It’d be interesting to see how the model flags accounts and whether the dataset includes explanations for predictions, like key behaviors or patterns. Plus, it opens the door for deeper discussions around ethics and privacy in spam detection.

Are you expecting any community contributions or feedback loops to enhance the model further? Or is this more about setting a precedent for transparency in the space? 🚀