AI RESEARCH
CART: Context-Anchored Recurrent Transformer -- A Parameter-Efficient Architecture with Learned Stability
arXiv CS.LG
•
ArXi:2606.01495v1 Announce Type: new We present CART (Context-Anchored Recurrent Transformer), a parameter-efficient language model that reuses a single shared core block R times across depth. Unlike prior looped transformers that recompute key-value tensors at every iteration, CART computes K and V once from a multi-layer prelude and has the recurrent core cross-attend to those frozen tensors via multi-head latent attention.