Payam Delgosha

About

Payam Delgosha is a researcher at the intersection of reinforcement learning, robotics, and large language models (LLMs), with a focus on automating complex skill acquisition. His most notable contribution is the introduction of **CurricuLLM**, a framework that leverages LLMs to automatically design task curricula for training robots. This work addresses a critical bottleneck in curriculum learning—the need for extensive human domain knowledge—by enabling machines to autonomously structure training sequences that progressively increase in difficulty. By bridging LLMs with reinforcement learning, Delgosha’s approach promises to accelerate the development of adaptive, real-world robotic systems. Though early in its citation trajectory, this 2025 paper has already garnered attention for its innovative fusion of language models and robotics. Delgosha’s research is particularly impactful for students and engineers seeking to reduce manual engineering in robot training, offering a scalable path toward more intelligent, self-directed learning agents. His work exemplifies how generative AI can reshape foundational RL paradigms, making complex skill acquisition more accessible and efficient.

Research Focus

Key Achievements

Top Papers

1. 1
   CurricuLLM: Automatic Task Curricula Design for Learning Complex Robot Skills Using Large Language Models

   1 citations · 2025

Key Collaborators

Contact & Links