AI Paper Atlas

Research map / Architectures and efficiency / Pruning and efficient tuning

HKAFER: Achieve Visual Parameter-Efficient Fine-Tuning via Heterogeneous Kronecker Adaptation for Facial Expression Recognition

Yu Gao, Haoyu Ji, Zhiyong Wang, Wenze Huang, Qian Dong, Zhihao Yang, Xueting Liu, Weihong Ren, Honghai Liu

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
use DBIR router instead of summing adapter outputs (ablation)HKAFER (DINOv2-ViT-L/14) · RAF-DBaccuracy (%): 92.28 -> 92.70Improved
use DBIR instead of Semantic Branch only (ablation)HKAFER (DINOv2-ViT-L/14) · RAF-DBaccuracy (%): 92.44 -> 92.70Improved
use DBIR instead of Relational Branch only (ablation)HKAFER (DINOv2-ViT-L/14) · ExpWaccuracy (%): 74.69 -> 74.81Improved
use heterogeneous Kronecker adapters (HeKA) instead of homogeneous (HoKA) (ablation)HKAFER (DINOv2-ViT-L/14) · RAF-DBaccuracy (%): 92.24 -> 92.70Improved
use HeKA instead of LoRA rank 32 (ablation)HKAFER (DINOv2-ViT-L/14) · RAF-DBaccuracy (%): 91.98 -> 92.70Improved
use HeKA instead of full fine-tuning (ablation)DINOv2-ViT-L/14 · RAF-DBaccuracy (%): 89.99 -> 92.70Improved
use HeKA instead of LoRA rank 32 (ablation)HKAFER (DINOv2-ViT-L/14) · ExpWaccuracy (%): 74.18 -> 74.81Improved
set number of adapters N to 4 rather than 1, 2 or 9 (ablation)HeKA · RAF-DB and ExpWaccuracy (%)Improved
use HKAFER vs CEPrompt under occlusionCLIP backbone · RAF-DB occlusionaccuracy (%): 90.13 -> 92.25Improved
use HKAFER vs JADFER in cross-dataset evaluationCLIP backbone · Cross-dataset mean (R to A, A to R)accuracy (%): 65.07 -> 70.24Improved
HKAFER on DINOv2 compared with larger-pretrained baselines on AffectNet and FERPlusDINOv2-ViT-L/14 · AffectNetaccuracy (%): 68.06 -> 67.66Hurt

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