ReToken: One Token to Improve Vision-Language Models for Visual Retrieval

In 30 Sekunden

Was passiert ist

ReToken adds a learnable token to vision-language models that selects relevant visual information from long contexts, improving retrieval performance while reducing memory demands.

Warum es zählt

Matters for engineers building or deploying vision-language systems that process long images or videos where GPU memory and distractor tokens limit performance.

Achtung

Training uses only small image-QA datasets; generalization to diverse long-context tasks and real-world deployment scenarios remains to be validated.