Early-Signal Detector

method · first seen 4 Oct 2026

prompt learning

Adapting vision-language models by learning continuous prompt embeddings instead of hand-written text.

Mentions per week

0219 May11 Aug10 NovPapersCodeDiscussion

PapersCodeDiscussion

Earliest sources

Latest sources

Mentioned together with

contrastive language-image pretrainingfew-shot anomaly detectiondiffusion model for prompt generationlow-rank adaptationadversarial triplet losszero-shot anomaly detectionaudio plugindomain generalizationmonocular depth estimationopen-vocabulary instance segmentation