Dawei

It's exciting to see renewed interest in weakly supervised learning as well as its promises and limitations when applied to LMs. 9/N cc @cs_dawei @andst_link

This is joint work with Xiaoyu Shen, Marius Mosbach (@mariusmosbach), Andreas Stephan (@andst_link ), and Dietrich Klakow. 🧵 12/N

We hope our findings and recommendations will spur more robust future work in WSL such that future methods are truly beneficial in realistic low-resource scenarios. 🙌🙌🙌 🧵11/N

📢 Check out our new #ACL2023 paper! "Weaker Than You Think: A Critical Look at Weakly Supervised Learning" ⚠ Want to apply weak supervision to solve your real-world tasks? Wait a second! ⚠ arxiv.org/abs/2305.17442 🧵 1/N

We will present our latest work on learning with noisily labeled data at #AAAI2021😊 If you are interested in distant supervision and noisy labels, we invite you to visit our poster this weekend or take a look into the paper arxiv.org/abs/2101.09763 @cs_dawei

"Neural Data-to-Text Generation with LM-based Text Augmentation" (Long Paper) Authors: Ernie Chang, Xiaoyu Shen, Dawei Zhu, Vera Demberg and Hui Su @erniecyc @cs_dawei