Search papers, labs, and topics across Lattice.
3
0
8
YouZhi-LLM achieves unprecedented concurrency and accuracy in financial LLMs by dramatically reducing KV-cache overhead, setting a new standard for deployment efficiency.
LiquidTAD achieves a remarkable 63% reduction in parameters while maintaining competitive performance, pushing the boundaries of efficiency in Temporal Action Detection.
Current multimodal retrieval systems fall flat when faced with realistic visual streams where context is distributed across time, motivating a new agentic paradigm for context-aware image retrieval.
