AI Paper Atlas

Research map / Vision-language and multimodal / Cross-modal learning

SDSA: Shallow-Deep Squeezing Adapter for Vision-Language Models

Hao Wang, Rui Zhu, Xiaoxu Li, Zhanyu Ma, Jing-Hao Xue

ECCV 2026 · Vision-language and multimodal › Cross-modal learning · approach: Contrastive vision-language alignment · 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
use SDSA instead of MMACLIP ViT-B/16 · Base-to-novel average (11 datasets)HM: 79.87 -> 80.96Improved
use SDSA instead of 2SFSCLIP ViT-B/16 · Base-to-novel average (11 datasets)Base accuracy: 85.55 -> 83.77Hurt
use SDSA instead of MMACLIP ViT-B/16 · Cross-dataset transfer from ImageNet (average)accuracy: 66.61 -> 67.32Improved
add low-rank transformation and cross-attention together (ablation)SDSA · Base-to-novel ablation (11-dataset average, HM)HM: 71.86 -> 80.96Improved
apply SDSA to all layers 1-12 instead of 9-12 (ablation)SDSA · Base-to-novel ablation (11-dataset average, HM)HM: 77.48 -> 80.96Improved
increase rank from 8 to 16 (ablation)SDSA · Base-to-novel ablation (11-dataset average, HM)HM: 80.51 -> 80.96Improved
increase rank from 16 to 64 (ablation)SDSA · Base-to-novel ablation (11-dataset average, HM)Novel accuracy: 78.34 -> 77.68Hurt
increase scaling factor alpha from 0.01 to 0.05 (ablation)SDSA · Base-to-novel ablation (11-dataset average, HM)HM: 77.10 -> 80.96Improved
increase scaling factor beta from 0.01 to 0.10 (ablation)SDSA · Base-to-novel ablation (11-dataset average, HM)Novel accuracy: 78.34 -> 76.54Hurt
increase scaling factor beta from 0.01 to 0.10 (ablation)SDSA · Base-to-novel ablation (11-dataset average, HM)Base accuracy: 83.77 -> 84.06Improved
add random token masking at ratio 0.10 versus no masking (ablation)SDSA · Base-to-novel ablation (11-dataset average, HM)HM: 80.91 -> 80.96No clear effect
raise mask ratio from 0.10 to 0.30 (ablation)SDSA · Base-to-novel ablation (11-dataset average, HM)HM: 80.96 -> 80.63Hurt

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