TMLR-Group-HF
/

Entropy-Qwen3-8B-Base-DAPO14k

Text Generation

text-generation-inference

Model card Files Files and versions

Entropy Minimization: Qwen3-8B-Base trained on DAPO-14k

This is the Qwen3-8B-Base model trained by Entropy Minimization using DAPO-14k training set, as described in the paper Co-rewarding: Stable Self-supervised RL for Eliciting Reasoning in Large Language Models.

For more details, installation, and usage examples, please refer to the official Github repository: https://github.com/tmlr-group/Co-rewarding.

Downloads last month: 4

Safetensors

Model size

8B params

Tensor type

BF16

·

Model tree for TMLR-Group-HF/Entropy-Qwen3-8B-Base-DAPO14k

Quantizations

Collection including TMLR-Group-HF/Entropy-Qwen3-8B-Base-DAPO14k

Co-rewarding

Co-rewarding is a novel self-supervised RL framework that improves training stability by seeking complementary supervision from another views. • 75 items • Updated Dec 21, 2025 • 1

Paper for TMLR-Group-HF/Entropy-Qwen3-8B-Base-DAPO14k

Co-Reward: Self-supervised Reinforcement Learning for Large Language Model Reasoning via Contrastive Agreement

Paper • 2508.00410 • Published Aug 1, 2025 • 1