Papers with text-based queries
BRIT: Bidirectional Retrieval over Unified Image-Text Graph (2025.findings-emnlp)
Copied to clipboard
| Challenge: | RAG is a promising technique to enhance the quality and relevance of responses generated by large language models. |
| Approach: | They propose a multi-modal RAG framework that unifies various text-image connections in a document into a graph and retrieves the texts and images as a query-specific sub-graph. |
| Outcome: | The proposed framework unifies various text-image connections into a multi-modal graph and retrieves the images and texts as a query-specific sub-graph. |