Exploring Multimodal Large Language Models ChatGPT-4 and Bard for Visual Complexity Evaluation of Mobile User Interfaces
Bibliographic record
Abstract
Generative large language models (LLM) are trained for performing natural language processing (NLP) tasks but are known to have emergent properties that can go beyond generating trained text-based language responses.Recently, LLMs have been further augmented with multimodal capabilities such as image annotations and analysis.In this study, we aimed to investigate LLMs in terms of perceptual visual complexity analysis ability through evaluating graphical user interfaces.For this purpose, visual complexity evaluation of user interfaces (UI), which is a non-trivial task, was addressed to explore the possible roles and capabilities of the LLMs in this task.ChatGPT-4 and Bard, two of the most advanced multi modal LLMs, were explored and a comparative evaluation was conducted.According to this exploration, the two LLMs were able to evaluate the visual complexity of different input user interfaces and rank these regarding to their visual complexities.Although LLMs ranking were mostly similar to each other, relatively high differences with the user evaluation-based rankings were observed.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.003 | 0.019 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.000 |
| Science and technology studies | 0.000 | 0.001 |
| Scholarly communication | 0.002 | 0.002 |
| Open science | 0.001 | 0.002 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.005 | 0.001 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".