AI Paper Atlas

Research map / Label-efficient and robust learning / Domain adaptation and generalization

Rapidly Deploying On-Device Eye Tracking by Distilling Visual Foundation Models

Cheng Jiang, Jogendra Nath Kundu, David Colmenares, Fengting Yang, Joseph Robinson, Yatong An, Ali Behrooz

ECCV 2026 · Label-efficient and robust learning › Domain adaptation and generalization · approach: Geometric alignment and synthetic-to-real adaptation · 1 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
fine-tune DINOv3 on synthetic data instead of linear probingDINOv3 ViT-B · Project Aria ET (2,222 participants)E50U50: 5.47 -> 2.01Improved
add self-distillation and unlabeled real data to synthetic fine-tuning (optimized VFM)DINOv3 ViT-B · Project Aria ET (2,222 participants)E50U50: 2.01 -> 1.33Improved
same, at the tailDINOv3 ViT-B · Project Aria ET (2,222 participants)E90U90: 12.11 -> 8.32Improved
use DARE-GRAM domain adaptation instead of the proposed teacher optimizationDINOv3 ViT-B · Project Aria ET (2,222 participants)E50U50: 2.01 -> 4.84Hurt
distill into on-device model with DistillGaze instead of synthetic supervisionFBNet 256K · Project Aria ET (2,222 participants)E50U50: 3.48 -> 1.44Improved
same, at the tailFBNet 256K · Project Aria ET (2,222 participants)E90U90: 14.84 -> 8.45Improved
add self-distillation with unlabeled real data to on-device model (no VFM)FBNet 256K · Project Aria ET (2,222 participants)E50U50: 3.48 -> 2.20Improved
add optimized VFM teacher on top of on-device self-distillationFBNet 256K · Project Aria ET (2,222 participants)E50U50: 2.20 -> 1.44Improved
use MSE regression loss instead of DINO loss for teacher (ablation)Optimized VFM · Project Aria ET (2,222 participants)E50U50: 1.50 -> 1.33Improved
add synthetic supervision to self-distillation and pseudo-labels (ablation)Optimized VFM · Project Aria ET (2,222 participants) (loss weight ablation)E50U50: 9.53 -> 1.33Improved
add loss weight scheduler (ablation)Optimized VFM · Project Aria ET (2,222 participants) (loss weight ablation)E50U50: 8.73 -> 1.33Improved
add self-distillation and pseudo-label losses to synthetic supervision (ablation)Optimized VFM · Project Aria ET (2,222 participants) (loss weight ablation)E50U50: 2.22 -> 1.33Improved

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