DIAL: GRPO on looped models - a omar81939 Collection

omar81939 's Collections

DIAL: GRPO on looped models

RL4RLM: Training Native Recursive Language Models

DIAL: GRPO on looped models

updated Aug 13

Random-depth SFT and depth-as-action GRPO checkpoints for Ouro-1.4B-Thinking.

omar81939/Ouro-1.4B-Thinking-depth-SFT

Text Generation • 1B • Updated Aug 13 • 29
omar81939/Ouro-1.4B-Thinking-depth-GRPO

Text Generation • 1B • Updated Aug 13 • 36