Text Generation
Transformers
Safetensors
qwen3
conversational
text-generation-inference
ystluffy commited on
Commit
a729945
·
verified ·
1 Parent(s): 8485b8e

Update README.md

Browse files

update the url of the technical report

Files changed (1) hide show
  1. README.md +5 -9
README.md CHANGED
@@ -13,24 +13,19 @@ datasets:
13
 
14
 
15
  [![GitHub](https://img.shields.io/badge/GitHub-DASD--Thinking-181717?logo=github&logoColor=white)](https://github.com/D2I-ai/dasd-thinking) 
16
- [![GitHub](https://img.shields.io/badge/GitHub-Technical--Report-181717?logo=github&logoColor=white)](https://github.com/D2I-ai/dasd-thinking/blob/main/dasd_technical_report.pdf) 
17
 
18
 
19
 
20
  [![Hugging Face](https://img.shields.io/badge/%F0%9F%A4%97%20Checkpoint-DASD--4B--Thinking-yellow)](https://huggingface.co/Alibaba-Apsara/DASD-4B-Thinking) 
21
- [![ModelScope Model](https://img.shields.io/badge/🤖%20Checkpoint-DASD--4B--Thinking-624aff)](https://www.modelscope.cn/models/Alibaba-Apsara/DASD-4B-Thinking) 
22
 
23
 
24
  [![Hugging Face](https://img.shields.io/badge/%F0%9F%A4%97%20Checkpoint-DASD--30B--A3B--Thinking--Preview-yellow)](https://huggingface.co/Alibaba-Apsara/DASD-30B-A3B-Thinking-Preview) 
25
- [![ModelScope Model](https://img.shields.io/badge/🤖%20Checkpoint-DASD--30B--A3B--Thinking--Preview-624aff)](https://www.modelscope.cn/models/Alibaba-Apsara/DASD-30B-A3B-Thinking-Preview) 
26
-
27
 
28
 
29
  [![Hugging Face](https://img.shields.io/badge/%F0%9F%A4%97%20Dataset-Superior--Reasoning--SFT--gpt--oss--120b-red)](https://huggingface.co/datasets/Alibaba-Apsara/Superior-Reasoning-SFT-gpt-oss-120b) 
30
- [![ModelScope Model](https://img.shields.io/badge/🤖%20Dataset-Superior--Reasoning--SFT--gpt--oss--120b-124aff)](https://www.modelscope.cn/datasets/Alibaba-Apsara/Superior-Reasoning-SFT-gpt-oss-120b) 
31
 
32
  [![Hugging Face](https://img.shields.io/badge/%F0%9F%A4%97%20Dataset-Superior--Reasoning--SFT--gpt--oss--120b--Logprob-red)](https://huggingface.co/datasets/Alibaba-Apsara/Superior-Reasoning-SFT-gpt-oss-120b-Logprob) 
33
- [![ModelScope Model](https://img.shields.io/badge/🤖%20Dataset-Superior--Reasoning--SFT--gpt--oss--120b--Logprob-124aff)](https://www.modelscope.cn/datasets/Alibaba-Apsara/Superior-Reasoning-SFT-gpt-oss-120b-Logprob) 
34
 
35
 
36
 
@@ -91,7 +86,7 @@ DASD-4B-Thinking democratizes the training recipe:
91
 
92
  ## ⚙️ Post-Training Pipeline
93
 
94
- DASD-Thinking introduces a new paradigm of **Distribution-Aligned Sequence Distillation**. This represents an enhanced sequence-level distillation pipeline that incorporates **Temperature-scheduled Learning**, **Divergence-aware Sampling**, and **Mixed-policy Distillation** , achieving efficient capability transfer with a minimal amount of data (**448K**). Please refer to our [report](https://github.com/D2I-ai/dasd-thinking/blob/main/dasd_technical_report.pdf) for more details.
95
 
96
  <div style="text-align: center;">
97
  <img src="assets/pipeline.jpg" alt="DASD-Thinking training pipeline" style="width: 90%;">
@@ -177,11 +172,12 @@ While DASD-4B-Thinking demonstrates remarkable performance across mathematical,
177
  DASD-Thinking is developed by Alibaba Cloud, as part of our mission to advance open, efficient, and trustworthy reasoning systems. If you find this work useful in your research or applications, please cite our technical report.
178
 
179
  ```bibtex
180
- @misc{yan2026dasd,
181
  title={Distribution-Aligned Sequence Distillation for Superior Long-CoT Reasoning},
182
  author={Yan, Shaotian and Liu, Kaiyuan and Shen, Chen and Wang, Bing and Fan, Sinan and Zhang, Jun and Wu, Yue and Wang, Zheng and Ye, Jieping},
183
  year={2026},
184
- url={https://github.com/D2I-ai/dasd-thinking/blob/main/dasd_technical_report.pdf}
 
185
  }
186
 
187
  @article{liu2025where,
 
13
 
14
 
15
  [![GitHub](https://img.shields.io/badge/GitHub-DASD--Thinking-181717?logo=github&logoColor=white)](https://github.com/D2I-ai/dasd-thinking)&#160;
16
+ <a href="https://arxiv.org/abs/2601.09088" target="_blank"><img src="https://img.shields.io/badge/Technical Report-b5212f.svg?logo=arxiv" height="21px"></a>
17
 
18
 
19
 
20
  [![Hugging Face](https://img.shields.io/badge/%F0%9F%A4%97%20Checkpoint-DASD--4B--Thinking-yellow)](https://huggingface.co/Alibaba-Apsara/DASD-4B-Thinking)&#160;
 
21
 
22
 
23
  [![Hugging Face](https://img.shields.io/badge/%F0%9F%A4%97%20Checkpoint-DASD--30B--A3B--Thinking--Preview-yellow)](https://huggingface.co/Alibaba-Apsara/DASD-30B-A3B-Thinking-Preview)&#160;
 
 
24
 
25
 
26
  [![Hugging Face](https://img.shields.io/badge/%F0%9F%A4%97%20Dataset-Superior--Reasoning--SFT--gpt--oss--120b-red)](https://huggingface.co/datasets/Alibaba-Apsara/Superior-Reasoning-SFT-gpt-oss-120b)&#160;
 
27
 
28
  [![Hugging Face](https://img.shields.io/badge/%F0%9F%A4%97%20Dataset-Superior--Reasoning--SFT--gpt--oss--120b--Logprob-red)](https://huggingface.co/datasets/Alibaba-Apsara/Superior-Reasoning-SFT-gpt-oss-120b-Logprob)&#160;
 
29
 
30
 
31
 
 
86
 
87
  ## ⚙️ Post-Training Pipeline
88
 
89
+ DASD-Thinking introduces a new paradigm of **Distribution-Aligned Sequence Distillation**. This represents an enhanced sequence-level distillation pipeline that incorporates **Temperature-scheduled Learning**, **Divergence-aware Sampling**, and **Mixed-policy Distillation** , achieving efficient capability transfer with a minimal amount of data (**448K**). Please refer to our [report](https://arxiv.org/abs/2601.09088) for more details.
90
 
91
  <div style="text-align: center;">
92
  <img src="assets/pipeline.jpg" alt="DASD-Thinking training pipeline" style="width: 90%;">
 
172
  DASD-Thinking is developed by Alibaba Cloud, as part of our mission to advance open, efficient, and trustworthy reasoning systems. If you find this work useful in your research or applications, please cite our technical report.
173
 
174
  ```bibtex
175
+ @article{yan2026dasd,
176
  title={Distribution-Aligned Sequence Distillation for Superior Long-CoT Reasoning},
177
  author={Yan, Shaotian and Liu, Kaiyuan and Shen, Chen and Wang, Bing and Fan, Sinan and Zhang, Jun and Wu, Yue and Wang, Zheng and Ye, Jieping},
178
  year={2026},
179
+ journal={arXiv preprint arXiv:2601.09088},
180
+ url={https://arxiv.org/abs/2601.09088}
181
  }
182
 
183
  @article{liu2025where,