Attending to Multimodal Generation One Token at a Time

Attending to Multimodal Generation One Token at a Time

Multimodal large language models exhibit distinct attention patterns during generation, with attention to visual and textual modalities shifting based on semantic requirements, and…

Hugging Face · Daily Papers ·Varun Gupta, Vineet Gandhi · ·▲ 6 upvotes

Este artigo está em destaque na seleção diária de papers do Hugging Face, curada pela comunidade de pesquisa em IA.

Autores: Varun Gupta, Vineet Gandhi, Makarand Tapaswi

  • 6 upvotes da comunidade
  • Temas: multimodal large language models, autoregressive generation, attention shifts, semantic role, causal attention blocking, cross-modal leakage

Resumo

Resumo original (em inglês), extraído do paper:

Multimodal large language models exhibit distinct attention patterns during generation, with attention to visual and textual modalities shifting based on semantic requirements, and these patterns can be leveraged to improve task performance through targeted interventions.

Onde ler

compartilhar: