Thinking with DistilQwen: A Tale of Four Distilled Reasoning and Reward Model Series
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2025
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866911247070396416 |
|---|---|
| author | Cai, Wenrui Wang, Chengyu Yan, Junbing Huang, Jun Fang, Xiangzhong |
| author_facet | Cai, Wenrui Wang, Chengyu Yan, Junbing Huang, Jun Fang, Xiangzhong |
| contents | Recently, the demand for small and efficient reasoning models to support real-world applications has driven the development of knowledge distillation techniques that balance reasoning performance and inference speed. In this paper, we further extend the DistilQwen model family, initialized from the Qwen models, by introducing four model series specifically designed to meet industrial requirements. The distilled model collection comprises: (1) slow-thinking models, optimized for reasoning tasks that require high accuracy; (2) two series of adaptive-thinking models, which dynamically adjust reasoning strategies based on input tasks to maximize efficiency across diverse scenarios; and (3) distilled reward models, which enable further reinforcement learning of reasoning models using distilled knowledge. Comprehensive evaluations across multiple benchmarks demonstrate both high inference efficiency and strong reasoning performance for these models, as well as the practical utility of distilled reward models. We further show that these models support industry practitioners by providing scalable training and inference functionalities on the Alibaba Cloud PAI (Platform for Artificial Intelligence) platform. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2511_01354 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Thinking with DistilQwen: A Tale of Four Distilled Reasoning and Reward Model Series Cai, Wenrui Wang, Chengyu Yan, Junbing Huang, Jun Fang, Xiangzhong Computation and Language Artificial Intelligence Recently, the demand for small and efficient reasoning models to support real-world applications has driven the development of knowledge distillation techniques that balance reasoning performance and inference speed. In this paper, we further extend the DistilQwen model family, initialized from the Qwen models, by introducing four model series specifically designed to meet industrial requirements. The distilled model collection comprises: (1) slow-thinking models, optimized for reasoning tasks that require high accuracy; (2) two series of adaptive-thinking models, which dynamically adjust reasoning strategies based on input tasks to maximize efficiency across diverse scenarios; and (3) distilled reward models, which enable further reinforcement learning of reasoning models using distilled knowledge. Comprehensive evaluations across multiple benchmarks demonstrate both high inference efficiency and strong reasoning performance for these models, as well as the practical utility of distilled reward models. We further show that these models support industry practitioners by providing scalable training and inference functionalities on the Alibaba Cloud PAI (Platform for Artificial Intelligence) platform. |
| title | Thinking with DistilQwen: A Tale of Four Distilled Reasoning and Reward Model Series |
| topic | Computation and Language Artificial Intelligence |
| url | https://arxiv.org/abs/2511.01354 |