Search papers, labs, and topics across Lattice.
Northeastern University
2
0
5
MemoryCard transforms long videos into coherent, topic-focused segments, boosting long-video QA accuracy by over 21% while maintaining visual-token efficiency.
Reasoning with LLMs just got a whole lot faster: MemoSight cuts KV cache footprint by 66% and speeds up inference by 1.56x without sacrificing CoT performance.
