AI Paper Atlas

Research map / Architectures and efficiency / Pruning and efficient tuning

Alfa: Attentive Low-Rank Filter Adaptation for Structure-Aware Cross-Domain Personalized Gaze Estimation

He-Yen Hsieh, Wei-Te Mark Ting, H. T. Kung

AAAI 2026 · Architectures and efficiency › Pruning and efficient tuning · approach: Low-rank compression and merging · 0 citations (OpenAlex)

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 attention over rank slices (ablation)Alfa · Gaze cross-domain avgangular gaze error (deg): 7.29 -> 5.72Improved
increase attention heads from 1 to 16 (ablation)Alfa · Gaze cross-domain avgangular gaze error (deg): 6.20 -> 5.72Improved
increase attention heads from 16 to 32 (ablation)Alfa · Gaze cross-domain avgangular gaze error (deg): 5.72 -> 5.91Hurt
use LoRA rank 8 instead of 4 (ablation)Alfa · Gaze cross-domain avgangular gaze error (deg): 5.84 -> 5.72Improved
use LoRA rank 16 instead of 8 (ablation)Alfa · Gaze cross-domain avgangular gaze error (deg): 5.72 -> 5.90Hurt
use SVD rank 64 instead of 32 (ablation)Alfa · Gaze cross-domain avgangular gaze error (deg): 6.69 -> 5.72Improved
use SVD rank 64 instead of 128 (ablation)Alfa · Gaze cross-domain avgangular gaze error (deg): 5.90 -> 5.72Improved
use 5 personal samples instead of 1 (ablation)Alfa · Gaze cross-domain avgangular gaze error (deg): 6.13 -> 5.72Improved
use 10 personal samples instead of 5 (ablation)Alfa · Gaze cross-domain avgangular gaze error (deg): 5.72 -> 5.70No clear effect
use Alfa instead of LoRA with same SVD-truncated weightsgaze model (SVD rank 64) · Gaze cross-domain avgangular gaze error (deg): 5.95 -> 5.72Improved
use Alfa instead of LoRA rank 128 on LLaDA-8B-InstructLLaDA-8B-Instruct · Countdown (0-shot, 128)accuracy: 20.3 -> 27.3Improved
use Alfa instead of LoRA rank 128 on LLaDA-8B-InstructLLaDA-8B-Instruct · Sudoku (0-shot, 128)accuracy: 16.5 -> 9.7Hurt

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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