Author

Priyatham Kattakinda

h-index 3 View on Semantic Scholar

We have 1 of 5 papers

We haven’t gathered this author’s papers yet. Follow them and we’ll fetch their work.

Not the right person? Other researchers publish under this name.

Jul 2026

Do Agent Optimizers Compound? A Continual-Learning Evaluation on Terminal-Bench 2.0

A two-phase continual-learning evaluation built from hard tasks in Terminal-Bench 2.0 found that optimization gains compounded only when regression control was built into the optimization loop, providing an inductive bias against shortcut solutions that fail to generalize.

Wenxiao Wang, Priyatham Kattakinda, S. Feizi · arXiv.org · 2 citations