LLM-FSM: Scaling Large Language Models for Finite-State Reasoning in RTL Code Generation
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2026
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866910014495522816 |
|---|---|
| author | Wu, Yuheng Gokmen, Berk Xie, Zhouhua Li, Peijing Trippel, Caroline Raina, Priyanka Tambe, Thierry |
| author_facet | Wu, Yuheng Gokmen, Berk Xie, Zhouhua Li, Peijing Trippel, Caroline Raina, Priyanka Tambe, Thierry |
| contents | Finite-state reasoning, the ability to understand and implement state-dependent behavior, is central to hardware design. In this paper, we present LLM-FSM, a benchmark that evaluates how well large language models (LLMs) can recover finite-state machine (FSM) behavior from natural-language specifications and translate it into correct register transfer-level (RTL) implementations. Unlike prior specification-to-RTL benchmarks that rely on manually constructed examples, LLM-FSM is built through a fully automated pipeline. LLM-FSM first constructs FSM with configurable state counts and constrained transition structures. It then prompts LLMs to express each FSM in a structured YAML format with an application context, and to further convert that YAML into a natural-language (NL) specification. From the same YAML, our pipeline synthesizes the reference RTL and testbench in a correct-by-construction manner. All 1,000 problems are verified using LLM-based and SAT-solver-based checks, with human review on a subset. Our experiments show that even the strongest LLMs exhibit sharply declining accuracy as FSM complexity increases. We further demonstrate that training-time scaling via supervised fine-tuning (SFT) generalizes effectively to out-of-distribution (OOD) tasks, while increasing test-time compute improves reasoning reliability. Finally, LLM-FSM remains extensible by allowing its FSM complexity to scale with future model capabilities. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2602_07032 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | LLM-FSM: Scaling Large Language Models for Finite-State Reasoning in RTL Code Generation Wu, Yuheng Gokmen, Berk Xie, Zhouhua Li, Peijing Trippel, Caroline Raina, Priyanka Tambe, Thierry Artificial Intelligence Hardware Architecture Computation and Language Finite-state reasoning, the ability to understand and implement state-dependent behavior, is central to hardware design. In this paper, we present LLM-FSM, a benchmark that evaluates how well large language models (LLMs) can recover finite-state machine (FSM) behavior from natural-language specifications and translate it into correct register transfer-level (RTL) implementations. Unlike prior specification-to-RTL benchmarks that rely on manually constructed examples, LLM-FSM is built through a fully automated pipeline. LLM-FSM first constructs FSM with configurable state counts and constrained transition structures. It then prompts LLMs to express each FSM in a structured YAML format with an application context, and to further convert that YAML into a natural-language (NL) specification. From the same YAML, our pipeline synthesizes the reference RTL and testbench in a correct-by-construction manner. All 1,000 problems are verified using LLM-based and SAT-solver-based checks, with human review on a subset. Our experiments show that even the strongest LLMs exhibit sharply declining accuracy as FSM complexity increases. We further demonstrate that training-time scaling via supervised fine-tuning (SFT) generalizes effectively to out-of-distribution (OOD) tasks, while increasing test-time compute improves reasoning reliability. Finally, LLM-FSM remains extensible by allowing its FSM complexity to scale with future model capabilities. |
| title | LLM-FSM: Scaling Large Language Models for Finite-State Reasoning in RTL Code Generation |
| topic | Artificial Intelligence Hardware Architecture Computation and Language |
| url | https://arxiv.org/abs/2602.07032 |