Spectral Scaling Laws in Language Models: How Effectively Do Feed-Forward Networks Use Their Latent Space?
Paper
•
2510.00537
•
Published
•
1
•
2
Nandan Kumar Jha
nandan523
AI & ML interests
Privacy and security of LLMs, and Cryptographically-secure machine learning
Recent Activity
authored
a paper
25 days ago
A Random Matrix Theory Perspective on the Learning Dynamics of
Multi-head Latent Attention
authored
a paper
25 days ago
Spectral Scaling Laws in Language Models: How Effectively Do
Feed-Forward Networks Use Their Latent Space?
commented on
a paper
26 days ago
Spectral Scaling Laws in Language Models: How Effectively Do
Feed-Forward Networks Use Their Latent Space?
Organizations
None yet
commented
a
paper
26 days ago
commented
4
papers
10 months ago
Entropy-Guided Attention for Private LLMs
Paper
•
2501.03489
•
Published
•
14
•
8
Entropy-Guided Attention for Private LLMs
Paper
•
2501.03489
•
Published
•
14
•
8
Entropy-Guided Attention for Private LLMs
Paper
•
2501.03489
•
Published
•
14
•
8
Entropy-Guided Attention for Private LLMs
Paper
•
2501.03489
•
Published
•
14
•
8