3 ms·
This paper notes that average output can be flawed...which is the case with all AI-generated content and why it's important to note the importance of cherry pic
by minimaxir 5y ago
This paper notes that average output can be flawed...which is the case with all AI-generated content and why it's important to note the importance of cherry picking/selection bias whenever posting it publicly.
Fortunately, AI Image generation has helped visually communicate how AI content generation isn't quite sci-fi magic where you always get what you want with zero ambiguity. Yet.
- theptip 5y agoIt’s an interesting point - cherry-picking muddies the water in an academic context. However it’s how these models will actually be used in the real world (with a human “editor” in the loop) so in some sense you care more about “best of N” performance. Or maybe even “best performance from someone that knows how to AI-whisper” since these are the new content creators that will produce the best output with these models (or their descendants). This is really a case where the real-world market will provide an assessment regardless of the sort of academic assessment of quality in the OP.