Training Effective and Efficient RLMs
Results and lessons from training small Recursive Language Models with reinforcement learning to improve accuracy while reducing subcall costs.
Occasional thoughts on things I find interesting.
Results and lessons from training small Recursive Language Models with reinforcement learning to improve accuracy while reducing subcall costs.
The introduction chapter of my thesis on training cost-aware Recursive Language Models with reinforcement learning.
Reading notes on RL for LLMs, LoRA, and related papers.