Architecture, Data, and Scale Don't Explain Universality in Vision Models
martin_hebart · x · 2026-09-25
Thread follow-up: universality was not explained by differences in architecture, training data, training objective, model size, or ImageNet performance.
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
- 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
- Claude Code autoresearch loop discovers jailbreaks beating 30+ GCG attacks, accepted at NeurIPS 2026 — maksym_andr · 2026-09-25