Search papers, labs, and topics across Lattice.
The University of New South Wales
3
0
5
3
MMEACR achieves significant performance improvements in visually grounded recommendations by effectively integrating multimodal memory and collaborative reasoning.
Semantic grounding, not token probability, is the key to better multimodal RAG.
Agentic RAG gets a check-up: Doctor-RAG surgically repairs failures in reasoning trajectories, boosting accuracy while slashing token consumption compared to brute-force reruns.
