Attention Manifolds: Steering or Blocking Language Models by Editing Learned B-Spline Surfaces

Naveen Mysore

In standard transformer attention, a source token sends the same value vector to every receiver. The query determines \emph{how much} to attend but not \emph{what} to extract. This work introduces \textbf{attention manifolds}: learned 2D B-spline surfaces $S_d(q_d, k_d)$ that modulate each value dimension based on the query-key interaction. Each surface is a tensor-product cubic B-spline initialized to zero, preserving pretrained behavior. Applied to LLaMA 3.2-1B-Instruct and 3B-Instruct, attention manifolds reduce WikiText-2 validation perplexity by 2--2.5 points with 0.3\% parameter overhead. Across 112 diverse prompts, surfaces change greedy-decoded output for 69\% (1B) to 83\% (3B) of cases, with the strongest effects on ambiguous and polysemous inputs (94--100\% change rate). The surfaces improve output quality: correcting factual errors (\emph{``the CAP theorem has three main components''} $\to$ \emph{``it is impossible to guarantee all three''}), increasing precision (\emph{``impossible to know certain properties''} $\to$ \emph{``impossible to know both position and momentum''}), and adding specificity (a generic quote $\to$ an attributed Saint Augustine citation,…