Downloads · 30 days
22
7% of all-time downloads
taki555/Qwen3-4B-Thinking-2507-Art
Qwen3-4B-Thinking-2507-Art is a text generation model from taki555. Use it when you need the model to write or continue text. It is set up for transformers. The card lists the license as apache-2.0.
This is the CoT efficient version of the Qwen3-4B-Thinking-2507 model, presented in the paper The Art of Efficient Reasoning: Data, Reward, and Optimization.
Downloads · 30 days
22
7% of all-time downloads
All-time downloads
336
Public
Parameters
4.4B
8.8 GB on disk
Likes
1
Public
Click a slice to open those files.
.safetensors8.8 GB · 100%
From the Hugging Face model README
This is the CoT efficient version of the Qwen3-4B-Thinking-2507 model, presented in the paper The Art of Efficient Reasoning: Data, Reward, and Optimization.
The model was trained on the DeepScaleR-Easy dataset to incentivize short yet accurate thinking trajectories.
Large Language Models (LLMs) consistently benefit from scaled Chain-of-Thought (CoT) reasoning, but also suffer from heavy computational overhead. This model addresses efficient reasoning by using a two-stage training paradigm: length adaptation and reasoning refinement. Through reward shaping with Reinforcement Learning (RL), the model is optimized to maintain high performance across a wide spectrum of token budgets while avoiding the "short-is-correct" trap.
For more details, please visit the Project Page.
@inproceedings{wu2026art,
title={The Art of Efficient Reasoning: Data, Reward, and Optimization},
author={Taiqiang Wu and Zenan Xu and Bo Zhou and Ngai Wong},
year={2026},
url={https://arxiv.org/pdf/2602.20945}
}