Search papers, labs, and topics across Lattice.
1
0
3
Explicitly teaching LVLMs to reason step-by-step with reinforcement learning unlocks state-of-the-art performance on multimodal object-entity relation extraction.
