1 Open Issue Need Help Last updated: Jun 27, 2025

Open Issues Need Help

View All on GitHub
AI/ML Advanced Reasoning Models

AI Summary: The task is to find and report the learning rate used to train the Qwen3-4B models in the POLARIS project, based on the provided project README and issue description. The response should be concise and address the user's question about the difficulty of improving post-trained models using GRPO.

Complexity: 2/5
good first issue

Scaling RL on advanced reasoning models

Python