🤔 How do we train LLMs on real-world tasks where it’s hard to define a single verifiable answer? Our work at Scale AI introduces Rubrics as Rewards (RaR) — a framework for on-policy post-training that uses structured, checklist-style rubrics as interpretable reward signals. 🧵

Unfortunately, I had to miss out on attending in person in Vienna, but glad to see the recognition. We need more research on understanding data and posttraining of LLMs. Always a pleasure working with Alan and Junmo Kang

Our team: MohammadHossein Rezaei (MohammadHossein Rezaei), Robert Vacareanu (Robert Vacareanu), Zihao Wang (Zihao Wang), Clinton Wang (Clinton Wang), Yunzhong He (Yunzhong), Feyza Akyürek (Afra Feyza Akyürek) Paper: arxiv.org/pdf/2510.07284