2 papers
cs.LG2026
Continuous Latent Contexts Enable Efficient Online Learning in Transformers
Emile Anand, Abdullah Ateyeh, Xinyuan Cao +1
Large language models (LLMs) exhibit a strong capacity for in-context learning: Given labeled examples, they can generate good predictions without parameter updates. However, many…
math.PR2025
The smallest singular value of inhomogenous random rectangular matrices
Max Dabagia, Manuel Fernandez
Let () be a random matrix with with independent entries that have mean 0 variance 1 and bounded moment. We show that the smallest s…