X-Intelligence 3.0: Training and Evaluating Reasoning LLM for Semiconductor Display
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866915403957010432 |
|---|---|
| author | Yan, Xiaolin Liu, Yangxing Zheng, Jiazhang Liu, Chi Du, Mingyu Chen, Caisheng Liu, Haoyang Ding, Ming Li, Yuan Liao, Qiuping Li, Linfeng Mei, Zhili Wan, Siyu Li, Li Zhong, Ruyi Yu, Jiangling Liu, Xule Hu, Huihui Yue, Jiameng Cheng, Ruohui Yang, Qi Wu, Liangqing Zhu, Ke Zhang, Chi Jing, Chufei Zhou, Yifan Liang, Yan Li, Dongdong Wang, Zhaohui Zhao, Bin Wu, Mingzhou Zhou, Mingzhong Du, Peng Liao, Zuomin Dai, Chao Liang, Pengfei Zhu, Xiaoguang Zhang, Yu Gu, Yu Pan, Kun Wu, Yuan Guan, Yanqing Wu, Shaojing Feng, Zikang Ma, Xianze Cheng, Peishan Jiang, Wenjuan Ba, Jing Yu, Huihao Hu, Zeping Xu, Yuan Liu, Zhiwei Wang, He Lin, Zhenguo Liu, Ming Meng, Yanhong |
| author_facet | Yan, Xiaolin Liu, Yangxing Zheng, Jiazhang Liu, Chi Du, Mingyu Chen, Caisheng Liu, Haoyang Ding, Ming Li, Yuan Liao, Qiuping Li, Linfeng Mei, Zhili Wan, Siyu Li, Li Zhong, Ruyi Yu, Jiangling Liu, Xule Hu, Huihui Yue, Jiameng Cheng, Ruohui Yang, Qi Wu, Liangqing Zhu, Ke Zhang, Chi Jing, Chufei Zhou, Yifan Liang, Yan Li, Dongdong Wang, Zhaohui Zhao, Bin Wu, Mingzhou Zhou, Mingzhong Du, Peng Liao, Zuomin Dai, Chao Liang, Pengfei Zhu, Xiaoguang Zhang, Yu Gu, Yu Pan, Kun Wu, Yuan Guan, Yanqing Wu, Shaojing Feng, Zikang Ma, Xianze Cheng, Peishan Jiang, Wenjuan Ba, Jing Yu, Huihao Hu, Zeping Xu, Yuan Liu, Zhiwei Wang, He Lin, Zhenguo Liu, Ming Meng, Yanhong |
| contents | Large language models (LLMs) have recently achieved significant advances in reasoning and demonstrated their advantages in solving challenging problems. Yet, their effectiveness in the semiconductor display industry remains limited due to a lack of domain-specific training and expertise. To bridge this gap, we present X-Intelligence 3.0, the first high-performance reasoning model specifically developed for the semiconductor display industry. This model is designed to deliver expert-level understanding and reasoning for the industry's complex challenges. Leveraging a carefully curated industry knowledge base, the model undergoes supervised fine-tuning and reinforcement learning to enhance its reasoning and comprehension capabilities. To further accelerate development, we implemented an automated evaluation framework that simulates expert-level assessments. We also integrated a domain-specific retrieval-augmented generation (RAG) mechanism, resulting in notable performance gains on benchmark datasets. Despite its relatively compact size of 32 billion parameters, X-Intelligence 3.0 outperforms SOTA DeepSeek-R1-671B across multiple evaluations. This demonstrates its exceptional efficiency and establishes it as a powerful solution to the longstanding reasoning challenges faced by the semiconductor display industry. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2507_14430 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | X-Intelligence 3.0: Training and Evaluating Reasoning LLM for Semiconductor Display Yan, Xiaolin Liu, Yangxing Zheng, Jiazhang Liu, Chi Du, Mingyu Chen, Caisheng Liu, Haoyang Ding, Ming Li, Yuan Liao, Qiuping Li, Linfeng Mei, Zhili Wan, Siyu Li, Li Zhong, Ruyi Yu, Jiangling Liu, Xule Hu, Huihui Yue, Jiameng Cheng, Ruohui Yang, Qi Wu, Liangqing Zhu, Ke Zhang, Chi Jing, Chufei Zhou, Yifan Liang, Yan Li, Dongdong Wang, Zhaohui Zhao, Bin Wu, Mingzhou Zhou, Mingzhong Du, Peng Liao, Zuomin Dai, Chao Liang, Pengfei Zhu, Xiaoguang Zhang, Yu Gu, Yu Pan, Kun Wu, Yuan Guan, Yanqing Wu, Shaojing Feng, Zikang Ma, Xianze Cheng, Peishan Jiang, Wenjuan Ba, Jing Yu, Huihao Hu, Zeping Xu, Yuan Liu, Zhiwei Wang, He Lin, Zhenguo Liu, Ming Meng, Yanhong Computation and Language Large language models (LLMs) have recently achieved significant advances in reasoning and demonstrated their advantages in solving challenging problems. Yet, their effectiveness in the semiconductor display industry remains limited due to a lack of domain-specific training and expertise. To bridge this gap, we present X-Intelligence 3.0, the first high-performance reasoning model specifically developed for the semiconductor display industry. This model is designed to deliver expert-level understanding and reasoning for the industry's complex challenges. Leveraging a carefully curated industry knowledge base, the model undergoes supervised fine-tuning and reinforcement learning to enhance its reasoning and comprehension capabilities. To further accelerate development, we implemented an automated evaluation framework that simulates expert-level assessments. We also integrated a domain-specific retrieval-augmented generation (RAG) mechanism, resulting in notable performance gains on benchmark datasets. Despite its relatively compact size of 32 billion parameters, X-Intelligence 3.0 outperforms SOTA DeepSeek-R1-671B across multiple evaluations. This demonstrates its exceptional efficiency and establishes it as a powerful solution to the longstanding reasoning challenges faced by the semiconductor display industry. |
| title | X-Intelligence 3.0: Training and Evaluating Reasoning LLM for Semiconductor Display |
| topic | Computation and Language |
| url | https://arxiv.org/abs/2507.14430 |