This AI Research Introduces Breakthrough Methods for Tailoring Language Models to Chip Design

ChipNeMo explores the utilisation of LLMs for industrial chip design, employing domain adaptation techniques rather than relying on off-the-shelf LLMs. These techniques involve custom tokenisation, domain-adaptive pretraining, supervised fine-tuning with domain-specific guidance, and domain-adapted retrieval models. The study evaluates these methods through three LLM applications in chip design, resulting in notable performance enhancements compared to […]

The post This AI Research Introduces Breakthrough Methods for Tailoring Language Models to Chip Design appeared first on MarkTechPost.

ai research ai shorts applications artificial intelligence chip chip design chipnemo computer vision design domain domain adaptation editors pick fine-tuning guidance industrial language language model language models large language model llm llm applications llms machine learning research retrieval staff study supervised fine-tuning tech news technology through tokenisation