JARVIS: A Multi-Agent Code Assistant for High-Quality EDA Script Generation
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_ | 1866915447868227584 |
|---|---|
| author | Pasandi, Ghasem Kunal, Kishor Tej, Varun Shah, Kunjal Sun, Hanfei Jain, Sumit Li, Chunhui Deng, Chenhui Ene, Teodor-Dumitru Ren, Haoxing Pratty, Sreedhar |
| author_facet | Pasandi, Ghasem Kunal, Kishor Tej, Varun Shah, Kunjal Sun, Hanfei Jain, Sumit Li, Chunhui Deng, Chenhui Ene, Teodor-Dumitru Ren, Haoxing Pratty, Sreedhar |
| contents | This paper presents JARVIS, a novel multi-agent framework that leverages Large Language Models (LLMs) and domain expertise to generate high-quality scripts for specialized Electronic Design Automation (EDA) tasks. By combining a domain-specific LLM trained with synthetically generated data, a custom compiler for structural verification, rule enforcement, code fixing capabilities, and advanced retrieval mechanisms, our approach achieves significant improvements over state-of-the-art domain-specific models. Our framework addresses the challenges of data scarcity and hallucination errors in LLMs, demonstrating the potential of LLMs in specialized engineering domains. We evaluate our framework on multiple benchmarks and show that it outperforms existing models in terms of accuracy and reliability. Our work sets a new precedent for the application of LLMs in EDA and paves the way for future innovations in this field. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2505_14978 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | JARVIS: A Multi-Agent Code Assistant for High-Quality EDA Script Generation Pasandi, Ghasem Kunal, Kishor Tej, Varun Shah, Kunjal Sun, Hanfei Jain, Sumit Li, Chunhui Deng, Chenhui Ene, Teodor-Dumitru Ren, Haoxing Pratty, Sreedhar Software Engineering Artificial Intelligence Machine Learning This paper presents JARVIS, a novel multi-agent framework that leverages Large Language Models (LLMs) and domain expertise to generate high-quality scripts for specialized Electronic Design Automation (EDA) tasks. By combining a domain-specific LLM trained with synthetically generated data, a custom compiler for structural verification, rule enforcement, code fixing capabilities, and advanced retrieval mechanisms, our approach achieves significant improvements over state-of-the-art domain-specific models. Our framework addresses the challenges of data scarcity and hallucination errors in LLMs, demonstrating the potential of LLMs in specialized engineering domains. We evaluate our framework on multiple benchmarks and show that it outperforms existing models in terms of accuracy and reliability. Our work sets a new precedent for the application of LLMs in EDA and paves the way for future innovations in this field. |
| title | JARVIS: A Multi-Agent Code Assistant for High-Quality EDA Script Generation |
| topic | Software Engineering Artificial Intelligence Machine Learning |
| url | https://arxiv.org/abs/2505.14978 |