DenseAMs show tradeoffs between entropy production, retrieval accuracy, and speed at intermediate loads, with a new failure mode in higher-order networks at finite temperature.
Hierarchical associative memory.arXiv preprint 2107.06446
4 Pith papers cite this work. Polarity classification is still indexing.
4
Pith papers citing it
citation-role summary
background 1
citation-polarity summary
years
2026 4verdicts
UNVERDICTED 4roles
background 1polarities
background 1representative citing papers
Uniform-based discrete diffusion models behave as associative memories that retrieve unseen data, with a dataset-size-driven memorization-to-generalization transition detectable via conditional entropy of token predictions.
Hierarchical Hopfield models retrieve concepts from noisy data via a strokes-concepts structure even without perfect stroke retrieval, as the second layer compensates for first-layer errors in both fixed- and variable-sized cases.
The paper reviews and extends energy-based dynamical models that use gradient flows and energy landscapes for neurocomputation, learning, and optimization tasks.
citing papers explorer
-
Stochastic Thermodynamics of Associative Memory
DenseAMs show tradeoffs between entropy production, retrieval accuracy, and speed at intermediate loads, with a new failure mode in higher-order networks at finite temperature.
-
Language Diffusion Models are Associative Memories Capable of Retrieving Unseen Data
Uniform-based discrete diffusion models behave as associative memories that retrieve unseen data, with a dataset-size-driven memorization-to-generalization transition detectable via conditional entropy of token predictions.
-
A mathematical analysis of hierarchical Hopfield models
Hierarchical Hopfield models retrieve concepts from noisy data via a strokes-concepts structure even without perfect stroke retrieval, as the second layer compensates for first-layer errors in both fixed- and variable-sized cases.
-
Energy-Based Dynamical Models for Neurocomputation, Learning, and Optimization
The paper reviews and extends energy-based dynamical models that use gradient flows and energy landscapes for neurocomputation, learning, and optimization tasks.