AI Paper Atlas

Research map / Architectures and efficiency / Pruning and efficient tuning

Memory-Efficient Transfer Learning with Fading Side Networks via Masked Dual Path Distillation

Yutong Zhang, Jiaxin Chen, Honglin Chen, Kaiqi Zheng, Shengcai Liao, Hanwen Zhong, Weixin Li, Yunhong Wang

CVPR 2026 · Architectures and efficiency › Pruning and efficient tuning · approach: Low-rank compression and merging · 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
add dual-path (logits plus feature) distillation instead of logits-based S to B only (ablation)MDPD (UniPT-based) · Flickr30K ITRRsum: 529.7 -> 538.1Improved
add dual-path distillation instead of feature-based B to S only (ablation)MDPD (UniPT-based) · Flickr30K ITRRsum: 531.5 -> 538.1Improved
use backbone-to-side feature direction instead of side-to-backbone (ablation)MDPD (UniPT-based) · Flickr30K ITRRsum: 518.2 -> 531.5Improved
fade the side network at inference (ablation)MDPD (UniPT-based) · Flickr30K ITRQPS: 7.98 -> 14.71Improved
fade the side network at inference (ablation)MDPD (UniPT-based) · Flickr30K ITRRsum: 541.9 -> 541.2Hurt
fade the side network at inference (ablation)MDPD (SHERL-based) · Flickr30K ITRRsum: 543.9 -> 543.0Hurt
use convolutional generation block instead of self-attention (ablation)MDPD (UniPT-based) · Flickr30K ITRRsum: 537.4 -> 541.2Improved
distill features across all layers instead of layer 1 only (ablation)MDPD (UniPT-based) · Flickr30K ITRRsum: 536.8 -> 541.2Improved
apply MDPD on top of UniPTUniPT · ITR BERT+BUTDQPS: 12.75 -> 24.05Improved
apply MDPD on top of SHERLSHERL · Flickr30K ITRQPS: 10.24 -> 14.83Improved
apply MDPD on top of SHERLSHERL · GLUE (T5-base)inference time (ms): 350.4 -> 234.8Improved
apply MDPD on top of SHERLSHERL · GLUE (T5-base)mean score: 84.3 -> 84.8Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 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