AI Paper Atlas

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

When the Teacher Has More Bits: Self-Teacher Latent Distillation for Learned Image Compression

Abdellah El Mennaoui, Joseph Meehan, Jean-Luc Dugelay, Ghalia Hemrit

ECCV 2026 · Label-efficient and robust learning › Knowledge distillation · approach: Few-shot and metric distillation · 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 self-teacher VNLD distillationHPCM · Tecnick BD-rate (LPIPS)BD-rate: 0 -> -1.80Improved
use variance normalization in the latent distillation loss (ablation)VNLD (HPCM) · HPCM BD-rate (LPIPS)BD-rate: -0.70 -> -1.60Improved
use variance normalization (ablation)VNLD (HPCM) · HPCM BD-rate (PSNR)BD-rate: -1.03 -> -1.12Improved
add VNLD vs MSE on BMSBMS · Kodak BD-rate (MS-SSIM)BD-rate: +0.4 (MSE) -> -0.5Improved
use a much higher-rate teacher (lambda 2) instead of adjacent-rate teachers (ablation)BMS student lambda 0.0018 · BMS RD lossRD loss: 0.3691 -> 0.3613Improved
use CRD distillation under frozen decoderSegPIC · CLIC Pro BD-rate (PSNR)BD-rate: 0 -> +4.90Hurt
use VNLD under frozen decoderSegPIC · Kodak/CLIC Pro/JPEG AI BD-rateBD-rate: -0.47, -1.60, -1.14 (PSNR)Improved
distillation weight beta above 5e-3 or below 5e-4 (ablation)VNLD · Kodak (BMS lvl6)RD performanceHurt
hierarchical or multi-target teacher supervision (ablation)BMS · KodakRD performanceNo clear effect
domain-specialize with VNLD vs RD-only fine-tuningSegPIC · Selfie BD-rate (PSNR)BD-rate: -4.21Improved
domain-specialize with VNLD on out-of-domain KodakSegPIC · Kodak BD-rate (PSNR)BD-rate: +0.5Hurt
VNLD vs MSE on HPCM Tecnick PSNRHPCM · Tecnick BD-rate (PSNR)BD-rate: -0.91 (MSE) -> -0.65Hurt

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