Reproduce “PromptIQ: Who Cares About Prompts? Let System Handle It -- A Component-Aware Framework for T2I Generation” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Nisan Chhetri, Arpan Sainju
PromptIQ evaluates text-to-image generations by segmenting subject components, captioning them, and comparing captions with predefined labels using CAS, then refining prompts through a user-feedback loop. In four vehicle examples, CAS was higher for modified prompts; for a car, it rose from 0.16 to 0.43 while CLIP shifted from 0.26 to 0.23. These examples do not establish correctness or user benefit.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.