Xiao Yan
7 indexed papers
Publications per year
Top categories
Frequent co-authors
Research Timeline
The paper proposes SemBugger, a polymorphic backdoor attack that uses intensity-based poisoning to achieve diverse malicious outcomes in Semantic Communication (SC) systems, alongside a provable defense mechanism.
The paper presents a novel technology that uses zero-knowledge proofs to formally verify a software system's correctness against a public specification without revealing the system's internal details.
The paper introduces Q-ALIGN DT, a novel framework that improves conditioned sequence models by enforcing alignment between the input return-to-go (RTG) signal and the output policy's expected Q-value, leading to superior policy controllability and performance.
The paper introduces AgentDoG 1.5, a lightweight and scalable alignment framework that significantly improves AI agent safety and security for complex, open-world agentic scenarios.
The paper introduces AgentDoG 1.5, a lightweight and scalable alignment framework that significantly improves AI agent safety and security for complex open-world agent deployments.
BenchEvolver introduces a solution-centric evolutionary framework to automatically transform saturated coding benchmarks into significantly harder, high-quality, and diverse evaluation suites.
The paper introduces MiCU, a domain-specific LLM that significantly improves smart home command understanding, especially for ambiguous commands, by synthesizing training data and optimizing the model for efficiency.
Papers
BenchEvolver: Frontier Task Synthesis via Solution-Centric Evolution
Yangzhen Wu, Aaron J. Li, Wenjie Ma, Li Cao +9 more
BenchEvolver introduces a solution-centric evolutionary framework to automatically transform saturated coding benchmarks into significantly harder, high-quality, and diverse evaluation suites.