Search papers, labs, and topics across Lattice.
The Hong Kong University of Science and Technology (Guangzhou)
3
0
5
Forget reward engineering: this work shows LLMs can self-evolve and outperform larger models by learning to explore and summarize new environments autonomously.
Polarization cues, often overlooked, can significantly boost camouflaged object detection by explicitly guiding RGB feature learning, leading to state-of-the-art performance.
Ditch unreliable object proposals: L2G-Det uses local patch matching to guide SAM for robust instance segmentation in cluttered, open-world scenes.
