New Study Examines Implicit Bias in State Space Models and Linear Autoregressive Training
A new arXiv paper analyzes how linear state space models and linear autoregressive sequence models, which reuse the same parameters across a sample-dependent number of steps, are shaped by implicit bias during training.
Read the full story at arXiv cs.LG ↗
Timeline · 1 report
- 2026-11-10 05:00 · arXiv cs.LG
Implicit Bias in State Space Models and Linear Autoregressive Training