AI Paper Atlas

Research map / Label-efficient and robust learning / Recognition and long-tail

DiT-Distill: Open-Set Fine-Grained Retrieval via Generative Curriculum Knowledge

Xin Jiang, Hao Tang, Meiqi Cao, Junyao Gao, Fei Shen, Zechao Li

CVPR 2026 · Label-efficient and robust learning › Recognition and long-tail · approach: Dense features and foundation priors

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 Generative Curriculum Distillation of DiT knowledge to ViT-B/16 student (ablation)ViT-B/16 · CUB-200-2011Recall@1: 77.4 -> 87.2Improved
add Generative Curriculum Distillation of DiT knowledge to ViT-B/16 student (ablation)ViT-B/16 · Stanford CarsRecall@1: 72.8 -> 91.4Improved
add generative infusion module with vanilla DiT teacher at inference (ablation)ViT-B/16 · CUB-200-2011Recall@1: 77.4 -> 85.3 (time 2.6ms -> 33.5ms)Improved
add conditional discrepancy refinement (CDR) to the DiT teacher (ablation)Teacher-assisted model · CUB-200-2011Recall@1: 85.3 -> 86.5Improved
add curriculum alignment loss to distill and discard DiT (ablation)Refined teacher-assisted model · CUB-200-2011Recall@1: 86.5 -> 87.2 (time 33.5ms -> 2.6ms)Improved
use 32 learnable queries instead of 16 or 64 (ablation)Generative Infusion Module · CUB-200-2011Recall@1: k=4 86.5, k=16 86.9, k=32 87.2, k=64 87.0Improved
use 3-layer bridging encoder instead of MLP (n=0) (ablation)Generative Infusion Module · CUB-200-2011Recall@1: 85.1 -> 87.2Improved
increase bridging encoder depth from 3 to 6 layers (ablation)Generative Infusion Module · CUB-200-2011Recall@1: 87.2 -> 86.7Hurt
add coarser timesteps to the curriculum (t4 only to t2-t4) (ablation)Curriculum alignment loss · CUB-200-2011Recall@1: 86.7 -> 87.2Improved
add noisiest timestep t1 to the curriculum (ablation)Curriculum alignment loss · CUB-200-2011Recall@1: 87.2 -> 87.2No clear effect
use manual bounding boxes instead of detector for CDR data (ablation)DiT-Distill · CUB-200-2011Recall@1: 87.2 -> 87.4No clear effect
set distillation weight alpha too low or too high versus 2.0 (ablation)DiT-Distill · CUB-200-2011Recall@1Hurt

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