Enregistré dans:
| Auteurs principaux: | , , , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2026
|
| Sujets: | |
| Accès en ligne: | https://arxiv.org/abs/2602.00704 |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866908801806893056 |
|---|---|
| author | Lyu, Hanqi Huang, Di Zhu, Yaoyu Liu, Kangcheng Dou, Bohan Li, Chongxiao Jin, Pengwei Cheng, Shuyao Zhang, Rui Du, Zidong Guo, Qi Hu, Xing Chen, Yunji |
| author_facet | Lyu, Hanqi Huang, Di Zhu, Yaoyu Liu, Kangcheng Dou, Bohan Li, Chongxiao Jin, Pengwei Cheng, Shuyao Zhang, Rui Du, Zidong Guo, Qi Hu, Xing Chen, Yunji |
| contents | The generation of Register-Transfer Level (RTL) code is a crucial yet labor-intensive step in digital hardware design, traditionally requiring engineers to manually translate complex specifications into thousands of lines of synthesizable Hardware Description Language (HDL) code. While Large Language Models (LLMs) have shown promise in automating this process, existing approaches-including fine-tuned domain-specific models and advanced agent-based systems-struggle to scale to industrial IP-level design tasks. We identify three key challenges: (1) handling long, highly detailed documents, where critical interface constraints become buried in unrelated submodule descriptions; (2) generating long RTL code, where both syntactic and semantic correctness degrade sharply with increasing output length; and (3) navigating the complex debugging cycles required for functional verification through simulation and waveform analysis. To overcome these challenges, we propose LocalV, a multi-agent framework that leverages information locality in modular hardware design. LocalV decomposes the long-document to long-code generation problem into a set of short-document, short-code tasks, enabling scalable generation and debugging. Specifically, LocalV integrates hierarchical document partitioning, task planning, localized code generation, interface-consistent merging, and AST-guided locality-aware debugging. Experiments on RealBench, an IP-level Verilog generation benchmark, demonstrate that LocalV substantially outperforms state-of-the-art (SOTA) LLMs and agents, achieving a pass rate of 45.0% compared to 21.6%. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2602_00704 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | LocalV: Exploiting Information Locality for IP-level Verilog Generation Lyu, Hanqi Huang, Di Zhu, Yaoyu Liu, Kangcheng Dou, Bohan Li, Chongxiao Jin, Pengwei Cheng, Shuyao Zhang, Rui Du, Zidong Guo, Qi Hu, Xing Chen, Yunji Machine Learning The generation of Register-Transfer Level (RTL) code is a crucial yet labor-intensive step in digital hardware design, traditionally requiring engineers to manually translate complex specifications into thousands of lines of synthesizable Hardware Description Language (HDL) code. While Large Language Models (LLMs) have shown promise in automating this process, existing approaches-including fine-tuned domain-specific models and advanced agent-based systems-struggle to scale to industrial IP-level design tasks. We identify three key challenges: (1) handling long, highly detailed documents, where critical interface constraints become buried in unrelated submodule descriptions; (2) generating long RTL code, where both syntactic and semantic correctness degrade sharply with increasing output length; and (3) navigating the complex debugging cycles required for functional verification through simulation and waveform analysis. To overcome these challenges, we propose LocalV, a multi-agent framework that leverages information locality in modular hardware design. LocalV decomposes the long-document to long-code generation problem into a set of short-document, short-code tasks, enabling scalable generation and debugging. Specifically, LocalV integrates hierarchical document partitioning, task planning, localized code generation, interface-consistent merging, and AST-guided locality-aware debugging. Experiments on RealBench, an IP-level Verilog generation benchmark, demonstrate that LocalV substantially outperforms state-of-the-art (SOTA) LLMs and agents, achieving a pass rate of 45.0% compared to 21.6%. |
| title | LocalV: Exploiting Information Locality for IP-level Verilog Generation |
| topic | Machine Learning |
| url | https://arxiv.org/abs/2602.00704 |