AI Paper Atlas

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

Advancing Textual Prompt Learning with Anchored Attributes

Zheng Li, Yibing Song, Ming-Ming Cheng, Xiang Li, Jian Yang

ICCV 2025 · Vision-language and multimodal › CLIP and zero-shot · approach: Prompt learning for CLIP · 25 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 ATPrompt attribute anchoringCoOp · base-to-novel average, 11 datasetsharmonic mean: 71.66 -> 74.65Improved
add ATPrompt attribute anchoringCoCoOp · base-to-novel average, 11 datasetsharmonic mean: 75.83 -> 77.95Improved
add ATPrompt attribute anchoringMaPLe · base-to-novel average, 11 datasetsharmonic mean: 78.55 -> 79.21Improved
add ATPrompt attribute anchoringDePT · base-to-novel average, 11 datasetsharmonic mean: 77.29 -> 78.45Improved
add ATPrompt attribute anchoringPromptKD · base-to-novel average, 11 datasetsharmonic mean: 83.73 -> 84.35Improved
add ATPrompt attribute anchoringCoOp · cross-dataset averageaccuracy: 63.88 -> 65.26Improved
add ATPrompt attribute anchoringMaPLe · domain generalization averageaccuracy: 60.27 -> 60.60Improved
place class token at end rather than front or middle (ablation)CoOp + ATPrompt · ImageNetharmonic mean: 73.20 -> 73.33Improved
retain hard and soft attribute tokens in deep layers instead of dropping and re-adding (ablation)MaPLe + ATPrompt · ImageNetharmonic mean: 73.31 -> 73.70Improved
use searched attributes instead of irrelevant attributes (ablation)CoOp + ATPrompt · 11-dataset averageharmonic mean: 73.59 -> 74.65Improved
use searched attributes instead of common attributes (ablation)CoOp + ATPrompt · 11-dataset averageharmonic mean: 74.16 -> 74.65Improved
reorder the attributes (ablation)CoOp + ATPrompt · ImageNetharmonic mean: 73.24 vs 73.33No 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