Author

Liangjie Zhao

Observed name: Zhao, Liangjie

Institutions: ByteDance (China), Adelaide University, The University of Adelaide

Observed institutions: ByteDance (China), Adelaide University, The University of Adelaide

OpenAlex ID: A5135271602

Practice Makes Unsafe: Skill Misevolution in Self-Improving LLM Agents

2026 · Mao, Xutao, Zhao, Liangjie, et al. · arXiv (Cornell University) · 9

Agents Don't Just Agree, They Remember: Benchmarking Persistent Sycophancy in Self-Improving Personal Agents

2026 · Mao, Xutao, Zhao, Liangjie, et al. · arXiv (Cornell University) · 6

UnifiedPlayers: Enhance Tool-Integrated Reasoning in Agentic Reinforcement Learning

2026 · Liao, Wenjie, Zhao, Liangjie, et al. · arXiv (Cornell University)

STARE: Step-wise Temporal Alignment and Red-teaming Engine for Multi-modal Toxicity Attack

2026 · Mao, Xutao, Zhao, Liangjie, et al. · arXiv (Cornell University)

Works count: 4

Citation count: 15

H-index: 2

I10-index: