RNNs Cannot Think What Transformers Think Cheaply. ICLR 2026 Proved the Gap Is Exponential.

For a decade, we asked if RNNs can represent what Transformers represent. We proved they can. We forgot to ask how expensively. That…