Search papers, labs, and topics across Lattice.
CASIA, Key Laboratory of Super Intelligent Security of Multi-Modal Information, Tencent Inc
2
0
4
Visual reranking and active rejection in MMAgent-R$^2$ significantly boost retrieval accuracy in challenging KB-VQA tasks, outperforming traditional methods.
Forget brute-force scaling: REVERSE shows that teaching an agent *how* to search and verify evidence lets a smaller model beat giants at image geo-localization.
