Provably Optimal Memory Capacity for Modern Hopfield Models: Transformer-Compatible Dense Associative Memories as Spherical Codes Jerry Y ao-Chieh Hu Dennis Wu

Nov-19-2025, 19:08:30 GMT–Neural Information Processing Systems

We study the optimal memorization capacity of modern Hopfield models and Kernelized Hopfield Models (KHMs), a transformer-compatible class of Dense Associative Memories.

large language model, machine learning, natural language, (19 more...)

Neural Information Processing Systems

Nov-19-2025, 19:08:30 GMT

Conferences PDF

Country:
- Asia > Middle East
  - Jordan (0.04)
- Europe > Austria
  - Vienna (0.04)
- North America > United States
  - Illinois > Cook County > Evanston (0.04)
- South America > Brazil (0.04)

Genre:
- Research Report
  - Experimental Study (1.00)
  - New Finding (1.00)

Industry:
- Health & Medicine > Therapeutic Area > Neurology (0.67)

Technology:
- Information Technology > Artificial Intelligence
  - Cognitive Science > Problem Solving (0.62)
  - Machine Learning
    - Neural Networks > Deep Learning (1.00)
    - Statistical Learning (0.93)
  - Natural Language > Large Language Model (1.00)
  - Representation & Reasoning > Optimization (0.67)
  - Systems & Languages > Programming Languages (0.62)
  - Vision (0.92)