Qwen-3-8B-Base

The 'Qwen3-8B-Base' dataset/benchmark is used to evaluate reinforcement learning-based post-training methods for large language models, focusing on the statistical properties of policy-gradient estimators and their optimization algorithms.

Papers using Qwen-3-8B-Base (4)

Clip Your Sequences Fairly: Enforcing Length Fairness for Sequence-Level RL2025 · 2 citesAgent0: Unleashing Self-Evolving Agents from Zero Data via Tool-Integrated Reasoning2025Asymmetric Proximal Policy Optimization: mini-critics boost LLM reasoning2025OBLR-PO: A Theoretical Framework for Stable Reinforcement Learning2025