mlsanityData to 7 Oct 2026

From Mirage to Grounding: Towards Reliable Multimodal Circuit-to-Verilog Code Generation

arXiv 2604.27969 ↗First posted 30 Apr 2026cs.SE cs.AIcited by 0

On the map: Multimodal LLMs / vision-language models → Chart understanding and anomaly detection

Abstract (arXiv)

Multimodal large language models (MLLMs) are increasingly used to translate visual artifacts into code, from UI mockups into HTML to scientific plots into Python scripts. A circuit diagram can be viewed as a visual domain-specific language for hardware: it encodes timing, topology, and bit level semantics that are invisible to casual inspection yet safety critical once fabricated in silicon. Translating such diagrams into register-transfer-level(RTL) code therefore represents an extreme reliability test for vision-to-code generation. We reveal a phenomenon we call Mirage: replacing a circuit diagram with a blank image leaves Pass@k unchanged or even higher, because models bypass the visual input and instead exploit identifier semantics in the module header to retrieve canonical RTL templates. This constitutes a new, highly covert class of defect in AI-assisted code generation that directly undermines MLLMs' trustworthiness. To quantify the effect, we construct C2VEVAL and evaluate eight MLLMs under a paired Normal/Anony protocol in which Anony mode anonymizes all identifiers in both the diagram and the module header; Anony-mode scores drop sharply across all models, confirming that high Normal-mode accuracy is largely a Mirage. We then propose VeriGround (4B), trained with identifier anonymization, refusal augmentation, and D-ORPO (Decision-Focused ORPO) preference alignment that up-weights pivotal generate-or-refuse tokens. VeriGround achieves Functional Pass@1 of 46.11%/42.51%(Normal/Anony) with a False Refusal Rate of only 1.20%/0.00%, while maintaining >92% Refusal Rate on blank images. With only 4B parameters, VeriGround performs on par with GPT-5.4 under Normal and significantly outperforms all baselines under Anony, confirming genuine visual grounding.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Placed in Chart understanding and anomaly detection, one of 16 directions in Multimodal LLMs / vision-language models. Builds on 33 of its 51 references that are in the map. First posted on arXiv 30 Apr 2026.

Builds on

33 of 51 references.

Closest ancestors · 1

Same direction, newest first

  1. 2025ChartCoder: Advancing Multimodal Large Language Model for Chart-to-Code Generation11 Jan

Other references · 32

  1. 2021LoRA: Low-Rank Adaptation of Large Language Modelscited by 14,084
  2. 2021Swin Transformer: Hierarchical Vision Transformer using Shifted Windowscited by 12,082
  3. 2024DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Modelscited by 8,851
  4. 2021Evaluating Large Language Models Trained on Codecited by 8,053
  5. 2023Direct Preference Optimization: Your Language Model is Secretly a Reward Modelcited by 7,916
  6. 2024GPT-4o System Cardcited by 4,467
  7. 2024Gemini 1.5: Unlocking multimodal understanding across millions of tokens of contextcited by 3,175
  8. 2024LLaVA-OneVision: Easy Visual Task Transfercited by 2,669
  9. 2023InternVL: Scaling up Vision Foundation Models and Aligning for Generic Visual-Linguistic Taskscited by 2,534
  10. 2025InternVL3.5: Advancing Open-Source Multimodal Models in Versatility, Reasoning, and Efficiencycited by 1,419
  11. 2023LLaVA-Med: Training a Large Language-and-Vision Assistant for Biomedicine in One Daycited by 1,287
  12. 2021Transformers in Vision: A Surveycited by 1,009
20 more
  1. 2023A Survey on Multimodal Large Language Modelscited by 898
  2. 2020A Survey on Visual Transformercited by 653
  3. 2024ORPO: Monolithic Preference Optimization without Reference Modelcited by 556
  4. 2023VerilogEval: Evaluating Large Language Models for Verilog Code Generationcited by 246
  5. 2023Multimodal Large Language Models: A Surveycited by 193
  6. 2023RTLLM: An Open-Source Benchmark for Design RTL Generation with Large Language Modelcited by 143
  7. 2026MIRAGE: The Illusion of Visual Understandingcited by 57
  8. 2024Interaction2Code: Benchmarking MLLM-based Interactive Webpage Code Generation from Interactive Prototypingcited by 50
  9. 2024Revisiting VerilogEval: A Year of Improvements in Large-Language Models for Hardware Code Generationcited by 45
  10. 2025A Survey of Research in Large Language Models for Electronic Design Automationcited by 38
  11. 2025QiMeng-CodeV-R1: Reasoning-Enhanced Verilog Generationcited by 34
  12. 2025OpenLLM-RTL: Open Dataset and Benchmark for LLM-Aided Design RTL Generationcited by 30
  13. 2025LLaVA-MORE: A Comparative Study of LLMs and Visual Backbones for Enhanced Visual Instruction Tuningcited by 27
  14. 2025UICopilot: Automating UI Synthesis via Hierarchical Code Generation from Webpage Designscited by 22
  15. 2025DSCodeBench: A Realistic Benchmark for Data Science Code Generationcited by 15
  16. 2024Prototype2Code: End-to-end Front-end Code Generation from UI Design Prototypescited by 10
  17. 2025Large Language Model for Verilog Code Generation: Literature Review and the Road Aheadcited by 7
  18. 2025ArchXBench: A Complex Digital Systems Benchmark Suite for LLM Driven RTL Synthesiscited by 6
  19. 2026Semantic Consensus Decoding: Backdoor Defense for Verilog Code Generationcited by 3
  20. 2026EGM: Efficient Visual Grounding Language Modelscited by 2

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2025Comprehensive Verilog Design Problems: A Next-Generation Benchmark Dataset for Evaluating Large Language Models and Agents on RTL Design and Verification
  2. 2025DeepV: A Model-Agnostic Retrieval-Augmented Framework for Verilog Code Generation with a High-Quality Knowledge Base
  3. 2024CraftRTL: High-quality Synthetic Data Generation for Verilog Code Models with Correct-by-Construction Non-Textual Representations and Targeted Code Repair
  4. 2026CASS-RTL: Correctness-Aware Subspace Steering for RTL Generation with LLMs
  5. 2026Beyond Pixels: Vector-to-Graph Transformation for Reliable Schematic Auditing
  6. 2026VeriGraphi: A Multi-Agent Framework of Hierarchical RTL Generation for Large Hardware Designs
  7. 2025VeriMind: Agentic LLM for Automated Verilog Generation with a Novel Evaluation Metric
  8. 2026TaoD2C-Bench: Benchmarking MLLMs for Industrial UI Code Generation Beyond Visual Fidelity