mlsanityData to 7 Oct 2026

MechVQA: Benchmarking and Enhancing Multimodal LLMs on Comprehensive Mechanical Drawing Understanding

arXiv 2605.30794 ↗First posted 29 May 2026cs.CV cs.AIcited by 3

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

Abstract (arXiv)

Multimodal Large Language Models (MLLMs) have demonstrated significant achievements in general visual question answering (VQA) tasks. However, they remain brittle on mechanical engineering drawings, where high annotation density and weak domain knowledge, compounded by unreliable spatial relation reasoning under strict projection rules and geometric constraints, make decisive cues easy to miss and frequently lead to wrong answers. To bridge this gap, we introduce the first comprehensive mechanical drawing understanding dataset, MechVQA, created through a semi-automated construction and quality-control pipeline. MechVQA contains 3.3k high-density pictures with 21K question-answer pairs, spanning 10 different fine-grained tasks across three capability levels: Recognition, Reasoning, and Judging, providing a testbed to evaluate and improve MLLM understanding on real-world mechanical drawings. On top of MechVQA, we then develop the MechVL model through a multi-stage training paradigm, building a strong domain-specialized baseline. Extensive experimental results demonstrate that MechVL outperforms the strongest closed-source baseline by 7.57 percentage points on the MechVQA total score, significantly enhancing mechanical drawing understanding ability and providing a reusable foundation for deploying MLLMs in mechanical design and inspection scenarios.

Key facts. Cited by 3 papers in the map (data through 7 Oct 2026). Placed in Chart understanding and anomaly detection, one of 16 directions in Multimodal LLMs / vision-language models. Builds on 32 of its 47 references that are in the map. First posted on arXiv 29 May 2026.

Builds on

32 of 47 references.

Closest ancestors · 3

Same direction, newest first

  1. 2026AECV-Bench: Benchmarking Multimodal Models on Architectural and Engineering Drawings Understanding8 Jan
  2. 2024EEE-Bench: A Comprehensive Multimodal Electrical And Electronics Engineering Benchmark3 Nov
  3. 2023From Concept to Manufacturing: Evaluating Vision-Language Models for Engineering Design21 Nov

Other references · 29

  1. 2021Learning Transferable Visual Models From Natural Language Supervisioncited by 33,456
  2. 2022Training language models to follow instructions with human feedbackcited by 15,712
  3. 2021LoRA: Low-Rank Adaptation of Large Language Modelscited by 14,084
  4. 2023Visual Instruction Tuningcited by 7,792
  5. 2019Parameter-Efficient Transfer Learning for NLPcited by 4,195
  6. 2024Gemini 1.5: Unlocking multimodal understanding across millions of tokens of contextcited by 3,175
  7. 2023MiniGPT-4: Enhancing Vision-Language Understanding with Advanced Large Language Modelscited by 2,689
  8. 2025DAPO: An Open-Source LLM Reinforcement Learning System at Scalecited by 2,678
  9. 2023Qwen-VL: A Versatile Vision-Language Model for Understanding, Localization, Text Reading, and Beyondcited by 2,179
  10. 2023MMBench: Is Your Multi-modal Model an All-around Player?cited by 2,105
  11. 2023Code Llama: Open Foundation Models for Codecited by 2,105
  12. 2025Gemma 3 Technical Reportcited by 1,686
17 more
  1. 2024LlamaFactory: Unified Efficient Fine-Tuning of 100+ Language Modelscited by 1,442
  2. 2025InternVL3.5: Advancing Open-Source Multimodal Models in Versatility, Reasoning, and Efficiencycited by 1,419
  3. 2024A Survey on LLM-as-a-Judgecited by 1,384
  4. 2024MiniCPM-V: A GPT-4V Level MLLM on Your Phonecited by 1,061
  5. 2025DeepSeek-V3.2: Pushing the Frontier of Open Large Language Modelscited by 755
  6. 2025Kimi K2: Open Agentic Intelligencecited by 523
  7. 2025GLM-4.5V and GLM-4.1V-Thinking: Towards Versatile Multimodal Reasoning with Scalable Reinforcement Learningcited by 439
  8. 2024MM-LLMs: Recent Advances in MultiModal Large Language Modelscited by 277
  9. 2024MT-Bench-101: A Fine-Grained Benchmark for Evaluating Large Language Models in Multi-Turn Dialoguescited by 247
  10. 2015Exploring Models and Data for Image Question Answeringcited by 139
  11. 2025MinerU2.5: A Decoupled Vision-Language Model for Efficient High-Resolution Document Parsingcited by 108
  12. 2022A survey of multimodal deep generative modelscited by 47
  13. 2025MicroVQA: A Multimodal Reasoning Benchmark for Microscopy-Based Scientific Researchcited by 30
  14. 2025From Drawings to Decisions: A Hybrid Vision-Language Framework for Parsing 2D Engineering Drawings into Structured Manufacturing Knowledgecited by 8
  15. 2025VaseVQA: Multimodal Agent and Benchmark for Ancient Greek Potterycited by 6
  16. 2025VaseVQA-3D: Benchmarking 3D VLMs on Ancient Greek Potterycited by 4
  17. 2025MECHBench: A Set of Black-Box Optimization Benchmarks originated from Structural Mechanicscited by 4

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2026: 3.

3 papers.

  1. 2026CrossProjection: Geometric Grounding Beyond Viewpoint Change in Architectural Drawingscited by 0
  2. 2026Wnuan: Staged Post-Training for Question Answering over Proprietary Enterprise Knowledgecited by 0
  3. 2026MechReason: Benchmarking Multi-Image Multi-Hop Reasoning in Mechanical Engineeringcited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2026DrawingVQA: A Real-World Benchmark for Multi-Depth Visual-Textual Reasoning on Construction Drawings
  2. 2024DesignQA: A Multimodal Benchmark for Evaluating Large Language Models' Understanding of Engineering Documentation
  3. 2026Enginuity: A Dataset and Benchmark for Vision-Language Understanding of Engineering Diagrams
  4. 2024Muffin or Chihuahua? Challenging Multimodal Large Language Models with Multipanel VQA
  5. 2025PhysToolBench: Benchmarking Physical Tool Understanding for MLLMs
  6. 2026MMArch: Benchmarking Multimodal Reasoning Grounded in Architectural Evidence
  7. 2026M3-VQA: A Benchmark for Multimodal, Multi-Entity, Multi-Hop Visual Question Answering
  8. 2025A Multi-Stage Hybrid Framework for Automated Interpretation of Multi-View Engineering Drawings Using Vision Language Model