ElectroVizQA
626 visual questions on digital-electronics circuit diagrams testing whether multimodal LLMs actually read schematics.
What it does and how it connects
ElectroVizQA is a benchmark dataset for evaluating multimodal large language models on digital-electronics visual question answering. It consists of curated circuit diagrams with associated questions and structured answers, designed to challenge both visual understanding and domain-specific reasoning.
The dataset comprises approximately 626 visual questions on digital-electronics circuit diagrams commonly found in undergraduate curricula, covering topics across digital electronics. The paper assesses how well multimodal models can understand and solve digital electronic circuit questions, and positions the dataset as the first of its kind tailored to the visual question answering task in digital electronics.
The repository holds the dataset structure of circuit diagrams in JPG format with VQA pairs per image, plus the paper citation. The paper is arXiv 2412.00102 by Meshram, Karthikeyan, Bhavya, and Bhat.
Alternatives
Compare side by side- MMCircuitEval
MMCircuitEval tests circuit knowledge across EDA stages with 3,614 Q and A pairs, while ElectroVizQA tests visual question answering on digital-electronics diagrams.
- EEE-Bench
EEE-Bench tests EE knowledge with 2,860 problems across 10 subdomains, while ElectroVizQA tests visual question answering on digital-electronics diagrams.
- PCB-QA
PCB-QA asks questions about real boards from their KiCad files, while ElectroVizQA tests visual question answering on digital-electronics diagrams.
- CircuitReason-1K
CircuitReason-1K tests long-horizon visual-to-symbolic circuit reasoning with worked solutions, while ElectroVizQA tests visual question answering on digital-electronics diagrams.
- OmniSch
OmniSch tests schematic-to-graph visual reasoning, while ElectroVizQA tests visual question answering on digital-electronics diagrams.
Sources
Not verified
- License: the repository API record lists no license and the README states none
- Pricing: none stated on the repo
- Real user voices (no real linked post read for this page)
- Related Error Index entries (no matching entry verified)
- A named individual builder beyond the ldong-ai repository owner
Last verified November 6, 2026. Listings are informational. The evidence score measures public, verifiable evidence and is not an endorsement or a compatibility claim.