HNC paper: shifting ViT weights 1.3% keeps predictions but scrambles representations
fatihdin4en · x · 2026-10-15
A new paper introduces Hessian Null Space Continuation (HNC), showing how much representational diversity surrounds a trained network. Moving a ViT's weights by just 1.3% — while preserving its predictions — makes its representations less similar to the original than a random-init ViT. The work joins a recent wave of 'diverse solutions' research: networks that perform the same task with very different internal representations.
Related event: HNC Method Reveals Massive Redundancy in Trained Neural Networks(2 posts)→
More from Research
- LLM4Code workshop at ICSE 2027 opens call for papers, due Nov 13, 2026 — LingmingZhang · 2026-10-15
- ELSA3D, a unified 3D generation and captioning model, hits SOTA at half the FLOPs, accepted at NeurIPS — CSProfKGD · 2026-10-15
- Moving ViT weights 1.3% without changing predictions scrambles representations, challenging Platonic hypothesis — neuroecology · 2026-10-15
- NeurIPS Spotlight: time-reversed LMs (TRLM) boost best-of-N reranking by up to 5% — LucaAmb · 2026-10-15
- Next-Concept Models Claim 31x Faster Text Generation Than Diffusion LMs — HazyResearch · 2026-10-15
- NCM iteratively refines its way to solving a sudoku puzzle — HazyResearch · 2026-10-15