AI Paper Atlas

Research map / Vision-language and multimodal / CLIP and zero-shot

AlignedNorm: Prompting Vision–Language Models via Coupled Prompt Field

Qi Ma, Chen-Yang Wang, Dehong Gao, Deng-Ping Fan

ICML 2026 · Vision-language and multimodal › CLIP and zero-shot · approach: Prompt learning for CLIP

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 post-projection norm alignment L_proj (ablation)MMRL++ · Base-to-new (11 datasets avg)HM: 80.71 -> 80.97Improved
add layer-wise token norm alignment L_token alone (ablation)MMRL++ · Base-to-new (11 datasets avg)HM: 80.71 -> 80.52Hurt
add both L_proj and L_token (ablation)MMRL++ · Base-to-new (11 datasets avg)HM: 80.71 -> 81.45Improved
use L1 loss for norm alignment instead of hinge loss (ablation)norm alignment · Base-to-new (11 datasets avg)HM: 80.55 -> 80.97Improved
use L1 loss instead of smooth L1 or Ring loss (ablation)norm alignment · Base-to-new (11 datasets avg)HM: 80.85/80.90 -> 80.97Improved
add AlignedNorm to MMRL++ for new-class accuracyMMRL++ · Base-to-new (11 datasets avg), new classesaccuracy (%): 76.48 -> 77.79Improved
add AlignedNorm (task-agnostic inference) vs MMRL++ with decoupled inferenceMMRL++ · Base-to-new (11 datasets avg)HM: 81.43 -> 81.45No clear effect
add AlignedNorm to MMRL++ w/o decoupling in cross-dataset transferMMRL++ · Cross-dataset (avg)accuracy (%): 67.47 -> 67.61Improved
add AlignedNorm to MMRL++ in low-shot regimeMMRL++ · Few-shot, 1 shotaccuracy (%): 72.57 -> 72.93Improved
add AlignedNorm to MMRL++ at 16 shotsMMRL++ · Few-shot, 16 shotsaccuracy (%): 84.07 -> 84.08No clear effect
add AlignedNorm to MMRL++ w/o decoupling in domain generalizationMMRL++ · Domain generalization (avg)accuracy (%): 60.93 -> 60.96No clear effect
add AlignedNorm training lossMMRL++ · training cost (ImageNet)training time (s): 0.357 -> 0.357No clear effect

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