AI Paper Atlas

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

Beyond Heuristic Prompting: A Concept-Guided Bayesian Framework for Zero-Shot Image Recognition

Hui Liu, Kecheng Chen, Jialiang Wang, Xianming Liu, Wenya Wang, Haoliang Li

CVPR 2026 · Vision-language and multimodal › CLIP and zero-shot · approach: Attribute-based zero-shot learning · 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 adaptive soft-trim likelihood (ablation)CGBC Prior (50 prompts) · 11 datasets avgtop-1 accuracy: 68.56 -> 69.25Improved
add adaptive soft-trim likelihood (ablation)CGBC Prior (16 prompts) · 11 datasets avgtop-1 accuracy: 68.35 -> 68.95Improved
use sigmoid soft-trim instead of hard-trim (ablation)CGBC · 11 datasets avg, 50 promptstop-1 accuracy: 69.07 -> 69.25Improved
use sigmoid soft-trim instead of confidence-based selection (ablation)CGBC · 11 datasets avg, 50 promptstop-1 accuracy: 68.61 -> 69.25Improved
compose atomic concepts with 'or' instead of embedding averaging (ablation)CGBC · 11 datasets avg, 50 promptstop-1 accuracy: 68.2 -> 69.3Improved
use larger concept generator (Turbo vs Mini vs Nano) (ablation)CGBC · 5-dataset avgtop-1 accuracy: 59.88 -> 60.17 -> 60.99Improved
increase hard-negative neighborhood from 5 to 20 (ablation)CGBC (GPT-4.1 Mini) · 5-dataset avgtop-1 accuracy: 59.84 -> 60.32Improved
increase hard-negative neighborhood from 20 to 30 (ablation)CGBC (GPT-4.1 Mini) · 5-dataset avgtop-1 accuracy: 60.32 -> 60.04Hurt
raise outlier threshold lambda from 1.5 to 2.5 (ablation)CGBC · 11 datasets avgtop-1 accuracy: 69.07 -> 69.25Mixed
use CLIP logit scale for sigmoid slope k instead of 0.25 (ablation)CGBC · 11 datasets avgtop-1 accuracy: 69.11 -> 69.25No clear effect
swap GPT-4.1 turbo for Gemini 2.5 Flash Lite concept generator (ablation)CGBC · 5-dataset avgtop-1 accuracy: 61.0 -> 59.9No clear effect
add multi-view entropy-filtered augmentation (CGBC + View)CGBC (1, 50) · ImageNet variants OOD averagetop-1 accuracy: 58.95 -> 63.36Improved

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