Poster · Sydney
Safe Actions Can Form Unsafe Traces: Benchmarking and Shielding Compositional Emergent Risk in AI Agents
- When
- Tue 8 Dec 2026, 10:00–13:00 AEDT
- Where
- Hall 1-4, ICC Sydney
Exhibition Centre, Level 1 - Session
- Sydney Poster Session 1
- Authors
- Zhongze Wu, Xiu Su, Hongyan Xu, Yichao Cao, Yueyi Luo, Jun Long
- Organisations
- Central South University
- QRS classification
- Agentic AI, Language models, Safety, fairness and privacy, AI, people and society