AI Paper Atlas

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

CAPT: Confusion-Aware Prompt Tuning for Reducing Vision-Language Misalignment

Maoyuan Shao, Yutong Gao, Xinyang Huang, Lijuan Sun, Guoshun Nan, Chuang Zhu

CVPR 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 MGDE to fuse semantic and sample confusion information (SEM+SAM -> full) (ablation)CAPT · Base-to-novel, 11 datasets averageharmonic mean (%): 73.86 -> 83.90Improved
add sample-level mining on top of semantic-only mining (ablation)CAPT · Base-to-novel, 11 datasets averageharmonic mean (%): 71.62 -> 73.86Improved
semantic-level mining alone (ablation)CAPT · Base-to-novel, 11 datasets averagebase accuracy (%): 80.69 vs 87.41 fullHurt
sample-level mining alone (ablation)CAPT · Base-to-novel, 11 datasets averagenovel accuracy (%): 63.41 vs 80.90 fullHurt
add confusion bank statistics to the confusion score (ablation)CAPT · Base-to-novel, 11 datasets averageharmonic mean (%): 82.62 -> 83.90Improved
use model's pseudo-GT instead of real ground-truth to select confusion pairs (ablation)CAPT · Base-to-novel, 11 datasets averageharmonic mean (%): 83.06 -> 83.90Improved
retrieve representative confusing samples from the bank instead of random samples (ablation)CAPT · Base-to-novel, 11 datasets averageharmonic mean (%): 83.04 -> 83.90Improved
use both global attention and local depthwise convolution in the Diff-Manner Adapter (ablation)CAPT · Grad-CAM qualitativequalitative attentionImproved
switch from PromptKD to CAPTPromptKD · Base-to-novel, 11 datasets averageharmonic mean (%): 83.73 -> 83.90Improved
switch from PromptKD to CAPTPromptKD · Base-to-novel FGVCAircraftharmonic mean (%): 45.17 -> 44.50Hurt
switch from TAC to CAPTTAC · Domain generalization ImageNet-Aaccuracy (%): 51.73 -> 57.73Improved
switch from PromptKD to CAPTPromptKD · Cross-dataset transfer, averageaccuracy (%): 71.33 -> 71.95Improved

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