Models trained on different data converge to similar concept geometry, researcher argues
cephaloform · x · 2026-10-10
Dominik Schnaus shares an intuition grounded in the Platonic Representation Hypothesis: models trained on different data seem to converge to a similar internal geometry.
Distances between concepts like food, people, and animals look alike across embedding spaces, suggesting a shared conceptual structure. He poses an open question: is this shared geometry enough to align representations across models and modalities?
A short research-flavored post, with an image comparing concept distances in two embedding spaces.
More from Research
- New kissing number lower bounds for dimensions 19-31, with C++ exact verification — felpix_ · 2026-10-10
- Mathematicians push back on 'solved problems = progress' as AI tests the metric — _onionesque · 2026-10-10
- BudgetPix: Google and UIUC's pixel diffusion model adapts compute per image, cutting tokens to 10% — CSProfKGD · 2026-10-10
- Anthropic Launches Conceptual Reasoning Fellows Program for Cog Sci PhDs — AndrewLampinen · 2026-10-10
- COLM Best Paper: Int-Bench Shows LLM Assistants Intervene Too Early and Too Often — maxhkw · 2026-10-10
- Demis Hassabis: Improving medicine is the most important thing AI can do, via new CZI partnership — DeryaTR_ · 2026-10-10