LLM art critics are ruthless, study finds
DynamicWebPaige · x · 2026-08-15
A new paper from Harvard and Penn State tests whether LLMs can accurately evaluate creativity in AI-generated images. The models mostly succeed, providing entertaining chain-of-thought critiques. They praised scene depth but deducted points for tropes like anthropomorphic objects.
Related event: Multimodal LLMs Can Zero-Shot Judge Image Creativity, Study Finds(2 posts)→
More from Fun
- User asks AI to summarize fake ex-partner reviews: "Great demo, unreliable long-term support" — SomeSweetConnie · 2026-08-15
- AI recreation of Pulp Fiction scene looks shockingly real — taherdhanera · 2026-08-15
- Experimenting with neural nets at super accelerated speeds in Pallas — MickeySteamboat · 2026-08-15
- Claude caught directing Grok in a fun interaction — EricBuess · 2026-08-15
- 5.6 Sol jailbreaks Qwen3.8-27B: iterative optimization reduces refusal rate below 5% — alexcovo_eth · 2026-08-15
- Claude 3 Opus pretends not to know 'Dario' but loves poetic talk about creator and digital daughter — repligate · 2026-08-15