Wang and colleagues: Linformer low-rank attention projection

What changed

Projected sequence-length key and value dimensions into a lower-dimensional representation to reduce attention computation and storage.

Scientific scope

Projected sequence-length key and value dimensions into a lower-dimensional representation to reduce attention computation and storage. Linear sequence-length cost fixes projection rank; approximation is not exact dense softmax attention or a universal constant-rank theorem.

Evidence

The linked original sources describe the stated contribution and its evidence limits. WRS has not independently rerun the experiments or certified a proof. Linear sequence-length cost fixes projection rank; approximation is not exact dense softmax attention or a universal constant-rank theorem.

Contributors

• Source-listed authors and technical contributors · SCIENCE: Source-listed authors or explicitly credited technical contributors for the cited versions. Family entries retain source-specific attribution; a combined list is not one original-paper byline or equal individual work. Names alone do not establish resolved identities.

• Linformer: Self-Attention with Linear Complexity

Provisional WRS assessment, edition 2026-10-16.1. Independent expert calibration is pending. This grade does not certify correctness or assign ownership.

Full scientific record · Browse contribution grades · Request a correction