Search papers, labs, and topics across Lattice.
Shanghai Artificial Intelligence Laboratory
4
0
6
SIVA-RL reveals that outcome-conditioned supervision can dramatically enhance multimodal reasoning performance, outperforming traditional methods across diverse benchmarks.
Self-evolving knowledge graphs can dramatically enhance multimodal reasoning by continuously adapting to new information and correcting errors in real-time.
Forgetting in universal segmentation models can be reduced to just 2.44% with a novel generative replay framework that synchronizes task relations.
LLMs are still far from being able to generate expert-level clinical guidelines, despite advances in deep research systems.
