Track papers across labs, topics, and trends.
Search papers, labs, and topics across Lattice.
4
0
7
5
A unified proactive generation framework enables LLMs to form reusable factual memories from streaming video without compromising real-time perception.
Most current robotic manipulation policies fail to retain crucial historical information, leading to significant performance drops in partially observable tasks, but a novel memory framework boosts success rates across diverse challenges.
VideoChat3 achieves unprecedented generalization in video understanding while maintaining high efficiency, outperforming larger models with just 4 billion parameters.
Current video LLMs falter when faced with the demands of real-time interaction, a gap RIVER Bench directly addresses by providing a challenging new evaluation framework.