Xi Yang

Current manipulation policies may excel at basic tasks but often fail to transfer learned skills to new contexts, exposing a significant gap in their real-world applicability.

Fictitious knowledge entries can safeguard RAG databases without compromising query integrity, achieving high detection rates with minimal interference.

LLMs can be easily misled by fabricated evidence, with even top-tier models failing to fully mitigate this vulnerability.

LLMs that ace general web browsing still fail miserably at autonomous scientific literature discovery, revealing a critical gap in research-oriented AI agent capabilities.

Cracking ancient languages like Oracle Bone Script just got easier: a new agent-driven Vision-Language Model leverages the compositional structure of characters to automate reasoning and improve interpretation accuracy.

Forget hand-crafting reward functions for your intelligent tutoring system; this apprenticeship learning approach learns evolving pedagogical strategies directly from expert student behavior.