RAM-Net: דגם סדרתי בזמן קווי עם גישה נבחרת דלילה

תקציר מקורי באנגליתarXiv:2602.11958v2 Announce Type: replace Abstract: Linear attention offers an efficient alternative to full attention with a fixed-size recurrent state. However, this state is shared by all tokens, so information from distinct tokens becomes superposed within it and produces inter-token interference that degrades long-range fine-grained recall. To address this issue, we propose RAM-Net, which replaces dense access to a shared state with sparse address-based access. RAM-Net organizes the recurrent state as a fixed-size array of independent slots and uses an Address Decoder that maps each key or query into a sparse address, selecting a small subset of slots to write to or read from at each step. This design directs tokens with non-overlapping addresses to disjoint slots, suppressing inter-t