QiMeng-CRUX: Narrowing the Gap Between Natural Language and Verilog via Core Refined Understanding eXpression for Circuit Design
Not on the map. Why?
Abstract (arXiv)
Large language models (LLMs) have shown promising capabilities in hardware description language (HDL) generation. However, existing approaches often rely on free-form natural language descriptions that are often ambiguous, redundant, and unstructured, which poses significant challenges for downstream Verilog code generation. We treat hardware code generation as a complex transformation from an open-ended natural language space to a domain-specific, highly constrained target space. To bridge this gap, we introduce Core Refined Understanding eXpression (CRUX), a structured intermediate space that captures the essential semantics of user intent while organizing the expression for precise Verilog code generation. We further design a two-stage training framework, comprising Joint Expression Modeling and Dual-Space Optimization, to enhance the quality of both CRUX and Verilog code. Experiments across multiple Verilog generation benchmarks demonstrate that our model, CRUX-V, achieves state-of-the-art performance among general models, particularly under challenging design tasks. Furthermore, the CRUX space proves transferable and beneficial when used as input prompts for other code models, highlighting its effectiveness in narrowing the gap between free-form natural language descriptions and precise Verilog generation.
Key facts. Cited by 1 paper in the map (data through 14 Oct 2026). Builds on 21 of its 33 references that are in the map. First posted on arXiv 25 Nov 2025.
Builds on
21 of 33 references.
References · 21
- 2017Proximal Policy Optimization Algorithmscited by 15,215
- 2024DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Modelscited by 9,379
- 2025DAPO: An Open-Source LLM Reinforcement Learning System at Scalecited by 2,810
- 2024LlamaFactory: Unified Efficient Fine-Tuning of 100+ Language Modelscited by 1,464
- 2024DeepSeek-Coder: When the Large Language Model Meets Programming -- The Rise of Code Intelligencecited by 1,181
- 2023VerilogEval: Evaluating Large Language Models for Verilog Code Generationcited by 252
- 2024BetterV: Controlled Verilog Generation with Discriminative Guidancecited by 100
- 2024StepCoder: Improve Code Generation with Reinforcement Learning from Compiler Feedbackcited by 77
- 2020DAVE: Deriving Automatically Verilog from Englishcited by 48
- 2024MG-Verilog: Multi-grained Dataset Towards Enhanced LLM-assisted Verilog Generationcited by 46
- 2024Revisiting VerilogEval: A Year of Improvements in Large-Language Models for Hardware Code Generationcited by 45
- 2024CraftRTL: High-quality Synthetic Data Generation for Verilog Code Models with Correct-by-Construction Non-Textual Representations and Targeted Code Repaircited by 42
9 more
- 2024AutoVCoder: A Systematic Framework for Automated Verilog Code Generation using LLMscited by 36
- 2025HaVen: Hallucination-Mitigated LLM for Verilog Code Generation Aligned with HDL Engineerscited by 25
- 2024Rome was Not Built in a Single Step: Hierarchical Prompting for LLM-based Chip Designcited by 24
- 2024Automatically Improving LLM-based Verilog Generation using EDA Tool Feedbackcited by 19
- 2025VeriReason: Reinforcement Learning with Testbench Feedback for Reasoning-Enhanced Verilog Generationcited by 16
- 2025Insights from Verification: Training a Verilog Generation LLM with Reinforcement Learning with Testbench Feedbackcited by 14
- 2025ReasoningV: Efficient Verilog Code Generation with Adaptive Hybrid Reasoning Modelcited by 12
- 2025VRank: Enhancing Verilog Code Generation from Large Language Models via Self-Consistencycited by 11
- 2025hdl2v: A Code Translation Dataset for Enhanced LLM Verilog Generationcited by 2
Built on by
Citing papers per year: 2026: 1.
1 paper.