This article examines the threat of backdoor attacks on language model agents.
― 5 min read
Yankai Lin
Research reveals significant security risks in chat models from backdoor attacks.
― 6 min read
New models enhance reasoning skills across various tasks, improving AI performance.
― 6 min read
Explores the challenges of supervising advanced AI models with weaker counterparts.
― 6 min read
GUICourse aims to improve interaction with digital interfaces through targeted datasets for GUI agents.
― 4 min read
Research highlights in-context learning abilities in large language models.
― 6 min read
Exploring the efficiency and adaptability of language models through modular design.
― 6 min read
TrendSim simulates social media trends to study misinformation effects.
― 7 min read