More Universal Dimensions in Vision Models Are More Semantic and Conceptual
martin_hebart · x · 2026-09-25
Thread follow-up: more universal dimensions were more semantic/conceptual, confirmed by categorization-based analyses — universality in penultimate layers reflects high-level content.
More from Research
- Coding Agents Beat Hand-Engineered Planners at Generalized TAMP, 56%-95% vs 47% — FBK-NLP · 2026-09-25
- SAE Latents Encode Part-of-Speech as Distributed Feature Groups, Not Atomic Features — colinglab · 2026-09-25
- The Gaussian is enough: Toyota study finds non-Gaussian priors don't help fine-tuning LBMs — _krishna_murthy · 2026-09-25
- Architecture, Data, and Scale Don't Explain Universality in Vision Models — martin_hebart · 2026-09-25
- 162 Vision Models Compared: NeurIPS Paper Finds Universal Representations Align With Human and Monkey Brains — martin_hebart · 2026-09-25
- Similarity-Based Representation Factorization: A General Method for Interpretable Dimensions — martin_hebart · 2026-09-25