AI Paper Atlas

Research map / Architectures and efficiency / Attention and transformers

Robust Filter Attention: Self-Attention as Precision-Weighted State Estimation

Peter Racioppo

ICML 2026 · Architectures and efficiency › Attention and transformers · approach: Sparse and structured attention

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add precision-weighted RFA attention instead of RoPERoPE · WikiText-103 (eval L=4096)test perplexity: 72.69 -> 38.46Improved
add spectral coupling of decay to head frequency (SC-RFA)RFA (M1) · WikiText-103 (eval L=4096)test perplexity: 38.46 -> 37.19Improved
add DLE-derived precision kernel (ablation)SC-RFA (M2) · WikiText-103 (eval L=4096)test perplexity: 62.83 -> 37.19Improved
add multiplicative precision term P_dt (ablation)SC-RFA (M2) · WikiText-103 (eval L=512)test perplexity: 27.65 -> 27.54Improved
add value rotation and counter-rotation (ablation)SC-RFA (M2) · WikiText-103 (eval L=4096)test perplexity: 463.29 -> 37.19Improved
add rotations (ablation)SC-RFA (M2) · WikiText-103 (eval L=512)test perplexity: 28.58 -> 27.54Improved
add rotations (ablation)SC-RFA (M2) · WikiText-103 (eval L=4096)test perplexity: 26.83 -> 37.19Hurt
use power-law robust weight instead of exponential weight (ablation)SC-RFA (M2) · WikiText-103 (eval L=4096)test perplexity: 33.51 -> 37.19Hurt
use power-law robust weight instead of exponential weight (ablation)SC-RFA (M2) · WikiText-103 (eval L=512)test perplexity: 27.98 -> 27.54Improved
add decay and uncertainty to pure rotation (ablation)Pure rotation (M2.6) · WikiText-103 (eval L=4096)test perplexity: 131.29 -> 37.19Improved
increase damping b from 0.05 to 5.0 (ablation)SC-RFA · WikiText-103 (eval L=4096)test perplexity: 37.19 -> 28.16Improved
increase damping b from 0.05 to 5.0 (ablation)SC-RFA · WikiText-103 (eval L=512)test perplexity: 27.54 -> 27.91Hurt

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy