Explaining Modern Gated-Linear RNNs via a Unified Implicit Attention Formulation

2025-01-22 · ICLR 2025 Poster · anchor

Findings