When Rule Learning Breaks: Diffusion Fails to Learn Parity of Many Bits

Dec 1, 2025·

Binxu Wang

Binxu Wang

,

Emma Lucia Byrnes Finn

,

Bingbin Liu

· 0 min read

Type

Conference paper

Publication

Workshop on Structured Probabilistic Inference & Generative Modeling, NeurIPS 2025 (oral)

Last updated on Dec 1, 2025

Diffusion Science of AI Generative Model

Binxu Wang

Authors

Research Fellow

← Neuronal Tuning Aligns Dynamically with Object- and Texture Manifolds across the Visual Hierarchy Jan 1, 2026

Feature accentuation along the encoding axes of IT neurons uncovers hidden differences in model-brain alignment May 12, 2025 →