RICA replaces ICA's global generative model with local Riemannian geometry, introducing a disentanglement tensor based on the Hessian of the log-likelihood and Ricci curvature to measure pointwise disentanglement, which recovers sources across manifolds in controlled tests.
Disentangling dense embeddings with sparse autoencoders
3 Pith papers cite this work. Polarity classification is still indexing.
3
Pith papers citing it
citation-role summary
background 1
citation-polarity summary
years
2026 3verdicts
UNVERDICTED 3roles
background 1polarities
background 1representative citing papers
Formalizes concept learning in sparse autoencoders as set alignment between human-defined and model-induced concepts, distinguishing detection, separation, and approximation with geometric conditions for neuron representation.
A computational framework identifies more coherent themes in free-text survey data on race, gender, and sexual orientation than previous methods, with applications for survey design, explaining variation, and detecting identity discordance.
citing papers explorer
-
Disentanglement Beyond Generative Models with Riemannian ICA
RICA replaces ICA's global generative model with local Riemannian geometry, introducing a disentanglement tensor based on the Hessian of the log-likelihood and Ricci curvature to measure pointwise disentanglement, which recovers sources across manifolds in controlled tests.
-
A Geometric View for Understanding Concept Learning and Neuron Interpretation in Sparse Autoencoders
Formalizes concept learning in sparse autoencoders as set alignment between human-defined and model-induced concepts, distinguishing detection, separation, and approximation with geometric conditions for neuron representation.
-
In your own words: computationally identifying interpretable themes in free-text survey data
A computational framework identifies more coherent themes in free-text survey data on race, gender, and sexual orientation than previous methods, with applications for survey design, explaining variation, and detecting identity discordance.