My favorite part - the classification challenge is that in the Year of Our Lord 2026, negative examples ("100% human text") are hard to get. But you can generate as many positive examples ("100% LLM text") as you want, so it's the perfect application for semi-supervised learning. So elegant 🙏

Show Post

New from @kevinren09.bsky.social @manishraghav.bsky.social @nkgarg.bsky.social: A crisp characterization of how (potentially adversarial) distribution shift is a major challenge for AI text detection, and how semi-supervised learning helps fix this.

Show Post

Paper: arxiv.org/abs/2606.25152 Thread:

Show Post

Framing fairness problems as wicked problems and borrowing critical pragmatism from planning theory is a great move. Congrats to all three of you!

Show Post

Excited to share my first PhD paper, “Hitting a Moving Target: Test-Time Adaptation for AI Text Detection under Continual Distribution Shift” (accepted to #NeurIPS2026)! We frame AI detection as a continual learning problem, robustifying models to shifting human/AI writing. (1/N)

Show Post