TMLR-Group-HF
/

Majority-Voting-Qwen3-4B-Base-OpenRS

Text Generation

text-generation-inference

Model card Files Files and versions

Majority-Voting: Qwen3-4B-Base trained on OpenRS

This is the Qwen3-4B-Base model trained by Majority-Voting using OpenRS training set. For more details on Co-rewarding, please refer to the paper Co-rewarding: Stable Self-supervised RL for Eliciting Reasoning in Large Language Models.

If you are interested in Co-rewarding, you can find more details on our Github Repo [https://github.com/tmlr-group/Co-rewarding].

Downloads last month: 11

Safetensors

Model size

4B params

Tensor type

BF16

·

Model tree for TMLR-Group-HF/Majority-Voting-Qwen3-4B-Base-OpenRS

Quantizations

1 model

Collection including TMLR-Group-HF/Majority-Voting-Qwen3-4B-Base-OpenRS

Co-rewarding

Co-rewarding is a novel self-supervised RL framework that improves training stability by seeking complementary supervision from another views. • 69 items • Updated Oct 3 • 1