Haijun Liu
1 indexed paper
Recent (6 mo)
1With code
0Influential cites
0Benchmarked
0Publications per year
126
Top categories
AI×1Crypto×1Software Eng.×1
Frequent co-authors
Research Timeline
2026
Inverting the Shield: Systematically Generating Safety Tests from Policy Specifications
The paper introduces POLARIS, a novel framework that systematically generates comprehensive and verifiable safety tests for LLMs by formalizing natural language policies into First-Order Logic and exploring the resulting Semantic Policy Graph.
Highlighted terms show continued research focus across papers
Papers
cs.AIcs.CRcs.SERecentMay 24, 2026
Inverting the Shield: Systematically Generating Safety Tests from Policy Specifications
Xiaoyue Lu, Xianglin Yang, Haijun Liu, Jiahao Liu +3 more
The paper introduces POLARIS, a novel framework that systematically generates comprehensive and verifiable safety tests for LLMs by formalizing natural language policies into First-Order Logic and exp…
View →