y0news
← Feed
Back to feed
🧠 AI🟢 BullishImportance 6/10

QA-Dragon: Query-Aware Dynamic RAG System for Knowledge-Intensive Visual Question Answering

arXiv – CS AI|Zhuohang Jiang, Pangjing Wu, Xu Yuan, Wenqi Fan, Qing Li|
🤖AI Summary

Researchers have developed QA-Dragon, a new Query-Aware Dynamic RAG System that significantly improves knowledge-intensive Visual Question Answering by combining text and image retrieval strategies. The system achieved substantial performance improvements of 5-6% across different tasks in the Meta CRAG-MM Challenge at KDD Cup 2025.

Key Takeaways
  • QA-Dragon addresses limitations of existing RAG methods that retrieve from either text or images in isolation.
  • The system introduces domain and search routers for dynamic optimal retrieval strategy selection.
  • It supports multimodal, multi-turn, and multi-hop reasoning for complex VQA tasks.
  • Performance improvements ranged from 5.03% to 6.35% across single-source, multi-source, and multi-turn tasks.
  • The framework was evaluated on the Meta CRAG-MM Challenge at KDD Cup 2025 with significant results.
Read Original →via arXiv – CS AI
Act on this with AI
Stay ahead of the market.
Connect your wallet to an AI agent. It reads balances, proposes swaps and bridges across 15 chains — you keep full control of your keys.
Connect Wallet to AI →How it works
Related Articles