TinyR1-32B-Preview: Boosting Accuracy with Branch-Merge Distillation

sunlin, Guangxiang Zhao, xiaoqi, wuyuhan, Weihong Lin, JosephZhu, Change Jia, Linglin zhang, Jinzhu Wu, Junfeng Ran, Saier Hu, Zihan Jiang, Antonio, Wenrui Liu, Bin Cui, Tong Yang, Xiangzheng Zhang

TinyR1-32B-Preview: Boosting Accuracy with Branch-Merge Distillation: 15 upvotes on Hugging Face Daily Papers, #14 of 24 papers on 2025-03-10. Day-by-day upvote history.

The challenge of reducing the size of Large Language Models (LLMs) while maintaining their performance has gained significant attention. However, existing methods, such as model distillation and transfer learning, often fail to achieve high accuracy. To address this limitation, we introduce the Branch-Merge distillation approach, which enhances model compression through two phases: (1) the Branch Phase, where knowledge from a large teacher model is selectively distilled into specialized student models via domain-specific supervised fine-tuning (SFT); And (2) the Merge Phase, where these student models are merged to enable cross-domain knowledge transfer and improve generalization. We validate our distillation approach using DeepSeek-R1 as the teacher and DeepSeek-R1-Distill-Qwen-32B as the student. The resulting merged model, TinyR1-32B-Preview, outperforms its counterpart DeepSeek-R1-Distill-Qwen-32B across multiple benchmarks, including Mathematics (+5.5 points), Coding (+4.4 points) and Science (+2.9 points), while achieving near-equal performance to DeepSeek-R1 on AIME 2024. The Branch-Merge distillation approach provides a scalable solution for creating smaller, high-performing LLMs with reduced computational cost and time.

Paper page on Hugging Face · arXiv

Data: hysts-bot-data/daily-papers-stats and the Daily Papers API. Open data: tardellirs/paper-pulse-data. Sister project: Model Pulse, the download history of every model on the Hub.