Expired50% confidenceFactTime unknown
Multimodal large language models such as GPT-4V and Claude can directly understand prototype images and identify UI elements like input fields, buttons, and dropdown menus
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
7/2/2026
First Seen
Valid until: 9/30/2026(expired)
Sources
Semi-AI Mode: A Pragmatic Approach to API Automation Testing Frameworks
bilibili阿里自动化测试
Related Claims
UnverifiedMultimodal large models (such as GPT-4V, Claude, etc.) can recognize hand-drawn sketches and understand their layout intent, converting them into structured HTML/CSS code.78% similarUnverifiedVision-language models (VLMs) such as GPT-4V, the Claude 3 series, and Gemini Ultra can understand interfaces at the semantic level rather than the pixel level, and can directly read webpage screenshots to identify button functions and form fields.76% similarVerifiedThe underlying architecture of GPT-Image-2 integrates the semantic understanding capabilities of large language models, enabling it to automatically convert natural language descriptions into visual generation instructions.75% similarVerifiedVision-language models such as GPT-4V and the Claude 3 series can process both image and text inputs simultaneously, supporting multimodal interaction.75% similarUnverifiedDesign-to-Code systems typically rely on multimodal large models such as GPT-4o and Claude 3.5 for structured understanding of interface images.74% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/59671API
curl https://kongchang.com/api/v1/knowledge/claims/59671MCP
get_claim(id=59671)