AI Paper Atlas

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

Prompt Tuning for CLIP on the Pretrained Manifold

Xi Yang, Yuanrong Xu, Weigang Zhang, Guangming Lu, David Zhang, Jie Wen

ICML 2026 · Vision-language and multimodal › CLIP and zero-shot · approach: Prompt learning for CLIP · 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 cosine consistency constraints (ablation)ManiPT with structural bias · Base-to-novel, 11 datasets averagenovel accuracy: 68.32 -> 77.58Improved
add structural bias (additive fusion with frozen features) (ablation)ManiPT with cosine consistency · Base-to-novel, 11 datasets averageharmonic mean: 79.54 -> 81.42Improved
add structural bias alone to baseline prompt tuning (ablation)baseline deep prompt tuning · Base-to-novel, 11 datasets averageharmonic mean: 72.82 -> 75.71Improved
add cosine consistency alone to baseline prompt tuning (ablation)baseline deep prompt tuning · Base-to-novel, 11 datasets averageharmonic mean: 72.82 -> 79.54Improved
add cosine consistency alone to baseline prompt tuning (ablation)baseline deep prompt tuning · Base-to-novel, 11 datasets averagebase accuracy: 82.91 -> 82.85No clear effect
add LLM-based knowledge enrichment (LLM description prototypes) (ablation)ManiPT · Base-to-novel, 11 datasets averageharmonic mean: 81.42 -> 82.09Improved
use ManiPT instead of CoPromptprompt tuning · Cross-dataset transfer, averageaccuracy: 66.99 -> 68.04Improved
use ManiPT instead of TACprompt tuning · Domain generalization, ImageNet variants averageaccuracy: 61.63 -> 61.88Improved
use ManiPT instead of MaPLe in drift and novel accuracydeep prompting · 8 datasets, PCA manifold driftmanifold driftImproved

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