Wenjie Qu

This work proposes a practical framework for verifiable DP training using CPU-side TEEs together with untrusted GPUs, and addresses a fundamental efficiency-security tension: training entirely inside a CPU TEE is too slow, while unrestricted GPU offloading can allow malicious deviations from DP.

Multimodal unlearning could revolutionize how we handle sensitive data in AI, enabling targeted removal without sacrificing model performance.

Claw-like agents are vulnerable to severe security breaches, with malicious plugins achieving a 100% success rate in attacks.

Sentence-level watermarks can now survive aggressive paraphrasing attacks like sentence splitting and merging, thanks to a new alignment-based approach.

Text watermarks can now survive even aggressive paragraph-level paraphrasing, thanks to a new self-anchoring technique that breaks the robustness-quality tradeoff.

Autonomous AI agents that can independently sustain and extend their operation are closer than we think, but raise thorny security and governance questions we need to address now.

Now you can audit black-box LLM APIs for cheating (model substitution, overbilling) with <1% overhead, using verifiable computation.