Skip to the Good Part: Representation Structure & Inference-Time Layer Skipping in Diffusion vs. Autoregressive LLMs
Paper
• 2603.07475 • Published
• 2
We’re scaling AI to create new possibilities.
Skip to the Good Part: Representation Structure & Inference-Time Layer Skipping in Diffusion vs. Autoregressive LLMs
On the "Induction Bias" in Sequence Models