A Princeton Researcher Proposes Recurrent Looped Transformer (RLT) that Carries Decoder State across Every Token, Fixing 96 Blocks per Token with Unbounded Temporal Depth

Princeton researcher Yifan Zhang published a technical report proposing a Recurrent Looped Transformer (RLT) that pairs a causal encoder with a recurrent decoder carrying its final hidden state and layerwise sliding-window attention cache across every prompt and response token without resetting at the serving boundary. The reference tied configuration uses 48 encoder and 48 decoder layers, executing 96 logical blocks per token while the state path grows to 48t decoder blocks after t tokens, and the design specifies hardware-aware execution around the recurrent core plus an exact current-policy RL replay contract. No code, weights, or measured results have been released.