AI Paper Atlas

Research map / Label-efficient and robust learning / Knowledge distillation

EdgeSAM: Prompt-In-the-Loop Distillation for SAM

Chong Zhou, Xiangtai Li, Chen Change Loy, Bo Dai

IJCV 2025 · Label-efficient and robust learning › Knowledge distillation · approach: Few-shot and metric distillation · 92 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 prompt-in-the-loop distillation to encoder-only KD (ablation)EdgeSAM (RepViT) · SA-1K center pointmIoU: 64.6 -> 67.5Improved
add prompt-in-the-loop distillation to encoder-only KD (ablation)EdgeSAM (RepViT) · SA-1K GT boxmIoU: 82.0 -> 83.0Improved
add prompt-in-the-loop KD to TinyViT backbone (ablation)TinyViT-5M · SA-1K center pointmIoU: 63.7 -> 68.0Improved
add prompt-in-the-loop KD to EfficientViT backbone (ablation)EfficientViT-B1 · SA-1K center pointmIoU: 60.9 -> 65.8Improved
add granularity-prior RPN (ablation)EdgeSAM · COCO center pointmIoU: 48.0 -> 54.3Improved
add granularity-prior RPN (ablation)EdgeSAM · 2080 Ti speedFPS: 164.3 -> 123.9Hurt
use RepViT-M1 backbone instead of TinyViT-5M (FPN alignment) (ablation)EdgeSAM encoder-only KD · SA-1KFPS: 114.2 -> 164.3Improved
two-stage training instead of joint training (ablation)EdgeSAM · SA-1K center pointmIoU: 63.2 -> 67.4Improved
supervise decoder with mask loss only instead of image feature loss (ablation)EdgeSAM · SA-1K GT boxmIoU: 71.6 -> 83.1Improved
add attention-map or IoU alignment loss to mask loss (ablation)EdgeSAM decoder · SA-1K GT boxmIoU: 83.1 -> 82.8 (attn), 82.5 (IoU)No clear effect
add one prompt sampling loop (ablation)EdgeSAM · SA-1K center +2 pt.mIoU: 72.4 -> 79.0Improved
add a second prompt sampling loop (ablation)EdgeSAM · SA-1K center +2 pt.mIoU: 79.0 -> 79.2No clear effect

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