Zero-Shot and Few-Shot Video Question Answering with Multi-Modal Prompts - Inria - Institut national de recherche en sciences et technologies du numérique Accéder directement au contenu
Communication Dans Un Congrès Année : 2023

Zero-Shot and Few-Shot Video Question Answering with Multi-Modal Prompts

Résumé

Recent vision-language models are driven by large-scale pretrained models. However, adapting pretrained models on limited data presents challenges such as overfitting, catastrophic forgetting, and the cross-modal gap between vision and language. We introduce a parameter-efficient method to address these challenges, combining multimodal prompt learning and a transformer-based mapping network, while keeping the pretrained models frozen. Our experiments on several video question answering benchmarks demonstrate the superiority of our approach in terms of performance and parameter efficiency on both zero-shot and few-shot settings. Our code is available at https: //engindeniz.github.io/vitis.
Fichier principal
Vignette du fichier
16.pdf (1.79 Mo) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)

Dates et versions

hal-04285294 , version 1 (14-11-2023)

Licence

Paternité

Identifiants

  • HAL Id : hal-04285294 , version 1

Citer

Deniz Engin, Yannis Avrithis. Zero-Shot and Few-Shot Video Question Answering with Multi-Modal Prompts. ICCV 2023 - International Conference on Computer Vision, Oct 2023, Paris, France. pp.1-7. ⟨hal-04285294⟩
13 Consultations
12 Téléchargements

Partager

Gmail Facebook X LinkedIn More