Excited to share our new preprint: "Causal Effect Estimation with Latent Textual Treatments" 🚀
We tackle a core challenge in text-based causal inference: how to perform valid experimentation with latent concepts as treatments.
arxiv.org/abs/2602.15730
1/8 🧵
Valid causal estimation is based on a novel residualization methodology. We provide theoretical and empirical results showing that our approach significantly mitigates the estimation bias found in naive methods.
7/8
It was a fantastic experience collaborating on this with Amar Venugopal, @jannspiess, and my supervisor @amir_feder.
Check out the full paper for more details and results: arxiv.org/abs/2602.15730
8/8