Search papers, labs, and topics across Lattice.
Peking University
2
0
5
Small models can now outperform large models in prospective memory tasks, achieving a groundbreaking 82.9% Set-F1 score with a simple, training-free approach.
A single malicious message can trigger a self-replicating worm, ClawWorm, that autonomously infects and propagates across entire LLM agent ecosystems, even surviving agent restarts.
