Yue Zhang
5 papers · Latest:
Usability as a Weapon: Attacking the Safety of LLM-Based Code Generation via Usability Requirements
This paper introduces UPAttack, demonstrating how usability requirements can force LLMs to generate insecure code, achieving up to 98.1% attack success.
When to Trust Imagination: Adaptive Action Execution for World Action Models
This paper introduces an adaptive execution method for World Action Models (WAMs) that verifies future predictions against reality, improving robotic manipulation efficiency and robustness.
SafeReview: Defending LLM-based Review Systems Against Adversarial Hidden Prompts
SafeReview defends LLM-based peer review systems against adversarial hidden prompts using a co-evolving generator-defender framework.
AgentDID: Trustless Identity Authentication for AI Agents
AgentDID provides a decentralized, trustless identity authentication and state verification framework for autonomous AI agents using DIDs and VCs.
SAGE: Signal-Amplified Guided Embeddings for LLM-based Vulnerability Detection
SAGE introduces Signal-Amplified Guided Embeddings to overcome "Signal Submersion" in LLM-based vulnerability detection, achieving SOTA performance.
📬 Weekly AI Paper Digest
Get the top 10 AI/ML arXiv papers from the week — summarized, scored, and delivered to your inbox every Monday.