ACE-RTL: When Agentic Context Evolution Meets RTL-Specialized LLMs
Fuente:
arXiv
Guardado en:
| Autores principales: | , , , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2026
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866912977125376000 |
|---|---|
| author | Deng, Chenhui Yu, Zhongzhi Liu, Guan-Ting Pinckney, Nathaniel Khailany, Brucek Ren, Haoxing |
| author_facet | Deng, Chenhui Yu, Zhongzhi Liu, Guan-Ting Pinckney, Nathaniel Khailany, Brucek Ren, Haoxing |
| contents | Recent advances in LLMs have sparked growing interest in applying them to hardware design automation, particularly for accurate RTL code generation. Prior efforts follow two largely independent paths: (i) training domain-adapted RTL models to internalize hardware semantics, (ii) developing agentic systems that leverage frontier generic LLMs guided by simulation feedback. However, these two paths exhibit complementary strengths and weaknesses. In this work, we present ACE-RTL that unifies both directions through Agentic Context Evolution (ACE). ACE-RTL integrates an RTL-specialized LLM, trained on a large-scale dataset of 1.7 million RTL samples, with a frontier reasoning LLM through three synergistic components: the generator, reflector, and coordinator. These components iteratively refine RTL code toward functional correctness. We further analyze a parallel scaling strategy that reduces wall-clock iterations to first success by exploring diverse debugging trajectories concurrently. On the CVDP benchmark, ACE-RTL achieves up to a 41.02% pass rate improvement over 14 competitive baselines. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2602_10218 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | ACE-RTL: When Agentic Context Evolution Meets RTL-Specialized LLMs Deng, Chenhui Yu, Zhongzhi Liu, Guan-Ting Pinckney, Nathaniel Khailany, Brucek Ren, Haoxing Hardware Architecture Machine Learning Recent advances in LLMs have sparked growing interest in applying them to hardware design automation, particularly for accurate RTL code generation. Prior efforts follow two largely independent paths: (i) training domain-adapted RTL models to internalize hardware semantics, (ii) developing agentic systems that leverage frontier generic LLMs guided by simulation feedback. However, these two paths exhibit complementary strengths and weaknesses. In this work, we present ACE-RTL that unifies both directions through Agentic Context Evolution (ACE). ACE-RTL integrates an RTL-specialized LLM, trained on a large-scale dataset of 1.7 million RTL samples, with a frontier reasoning LLM through three synergistic components: the generator, reflector, and coordinator. These components iteratively refine RTL code toward functional correctness. We further analyze a parallel scaling strategy that reduces wall-clock iterations to first success by exploring diverse debugging trajectories concurrently. On the CVDP benchmark, ACE-RTL achieves up to a 41.02% pass rate improvement over 14 competitive baselines. |
| title | ACE-RTL: When Agentic Context Evolution Meets RTL-Specialized LLMs |
| topic | Hardware Architecture Machine Learning |
| url | https://arxiv.org/abs/2602.10218 |