Exploring the Capabilities of Vision-Language Models to Detect Visual Bugs in HTML5 Applications (v2)

Data for the replication package for the paper "Exploring the Capabilities of VLMs to Detect Visual Bugs in HTML5 <canvas> Applications" (v2) "Exploring the Capabilities of VLMs to Detect Visual Bugs in HTML5 <canvas> Applications" by Finlay Macklon, Cedric Boucher and Cor-Paul Bezemer Paper pre-print: ... Code on GitHub: https://github.com/asgaardlab/vlm_canvas_bugs-code Repositories containing <canvas> applications: vlm-canvas-bugs-external-repos GitHub account See the following link on instructions for how to use this data in the replication package: https://github.com/asgaardlab/vlm_canvas_bugs-code Directory structure: Data ├── 1a-Collecting_Canvas_Applications ├── 1b-Creating_E2E_Test_Scripts (empty) ├── 1c-Injecting_Visual_Bugs ├── 1d-Collecting_Screenshots ├── 2-Experiments └── 3-Results