AnalogSeeker: An Open-source Foundation Language Model for Analog Circuit Design
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , , , , , , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2025
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866914138501939200 |
|---|---|
| author | Chen, Zihao Zhuang, Ji Shen, Jinyi Ke, Xiaoyue Yang, Xinyi Zhou, Mingjie Du, Zhuoyao Yan, Xu Wu, Zhouyang Xu, Zhenyu Huang, Jiangli Shang, Li Zeng, Xuan Yang, Fan |
| author_facet | Chen, Zihao Zhuang, Ji Shen, Jinyi Ke, Xiaoyue Yang, Xinyi Zhou, Mingjie Du, Zhuoyao Yan, Xu Wu, Zhouyang Xu, Zhenyu Huang, Jiangli Shang, Li Zeng, Xuan Yang, Fan |
| contents | In this paper, we propose AnalogSeeker, an effort toward an open-source foundation language model for analog circuit design, with the aim of integrating domain knowledge and giving design assistance. To overcome the scarcity of data in this field, we employ a corpus collection strategy based on the domain knowledge framework of analog circuits. High-quality, accessible textbooks across relevant subfields are systematically curated and cleaned into a textual domain corpus. To address the complexity of knowledge of analog circuits, we introduce a granular domain knowledge distillation method. Raw, unlabeled domain corpus is decomposed into typical, granular learning nodes, where a multi-agent framework distills implicit knowledge embedded in unstructured text into question-answer data pairs with detailed reasoning processes, yielding a fine-grained, learnable dataset for fine-tuning. To address the unexplored challenges in training analog circuit foundation models, we explore and share our training methods through both theoretical analysis and experimental validation. We finally establish a fine-tuning-centric training paradigm, customizing and implementing a neighborhood self-constrained supervised fine-tuning algorithm. This approach enhances training outcomes by constraining the perturbation magnitude between the model's output distributions before and after training. In practice, we train the Qwen2.5-32B-Instruct model to obtain AnalogSeeker, which achieves 85.04% accuracy on AMSBench-TQA, the analog circuit knowledge evaluation benchmark, with a 15.67% point improvement over the original model and is competitive with mainstream commercial models. Furthermore, AnalogSeeker also shows effectiveness in the downstream operational amplifier design task. AnalogSeeker is open-sourced at https://huggingface.co/analogllm/analogseeker for research use. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2508_10409 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | AnalogSeeker: An Open-source Foundation Language Model for Analog Circuit Design Chen, Zihao Zhuang, Ji Shen, Jinyi Ke, Xiaoyue Yang, Xinyi Zhou, Mingjie Du, Zhuoyao Yan, Xu Wu, Zhouyang Xu, Zhenyu Huang, Jiangli Shang, Li Zeng, Xuan Yang, Fan Hardware Architecture Artificial Intelligence In this paper, we propose AnalogSeeker, an effort toward an open-source foundation language model for analog circuit design, with the aim of integrating domain knowledge and giving design assistance. To overcome the scarcity of data in this field, we employ a corpus collection strategy based on the domain knowledge framework of analog circuits. High-quality, accessible textbooks across relevant subfields are systematically curated and cleaned into a textual domain corpus. To address the complexity of knowledge of analog circuits, we introduce a granular domain knowledge distillation method. Raw, unlabeled domain corpus is decomposed into typical, granular learning nodes, where a multi-agent framework distills implicit knowledge embedded in unstructured text into question-answer data pairs with detailed reasoning processes, yielding a fine-grained, learnable dataset for fine-tuning. To address the unexplored challenges in training analog circuit foundation models, we explore and share our training methods through both theoretical analysis and experimental validation. We finally establish a fine-tuning-centric training paradigm, customizing and implementing a neighborhood self-constrained supervised fine-tuning algorithm. This approach enhances training outcomes by constraining the perturbation magnitude between the model's output distributions before and after training. In practice, we train the Qwen2.5-32B-Instruct model to obtain AnalogSeeker, which achieves 85.04% accuracy on AMSBench-TQA, the analog circuit knowledge evaluation benchmark, with a 15.67% point improvement over the original model and is competitive with mainstream commercial models. Furthermore, AnalogSeeker also shows effectiveness in the downstream operational amplifier design task. AnalogSeeker is open-sourced at https://huggingface.co/analogllm/analogseeker for research use. |
| title | AnalogSeeker: An Open-source Foundation Language Model for Analog Circuit Design |
| topic | Hardware Architecture Artificial Intelligence |
| url | https://arxiv.org/abs/2508.10409 |