Author

Jinghan Xu

Observed name: Xu, Jinghan

Institutions: Nankai University, Fudan University

Observed institutions: Nankai University, Fudan University

OpenAlex ID: A5144248359

Beyond Single-Use Tokens: Durable Authorization State for Replay-Resistant LLM Agent Actions

2026 · Xu, Jinghan, Fan, Longze, et al. · arXiv (Cornell University) · 4

Memory Provenance Laundering in LLM Agents: A Non-Amplification Firewall for Persistent Memory

2026 · Xu, Jinghan, Xiao, Yiyong, et al. · arXiv (Cornell University) · 2

Beyond Skill Evolution: Self-Evolving Context Management Policies for Long-Horizon Agent Harnesses

2026 · Li, Weiyuan, Xu, Jinghan, et al. · arXiv (Cornell University)

EvoRS: On-Policy Self-Evolution of Reward Systems for Open-Ended Reinforcement Learning

2026 · Li, Weiyuan, Chen, Aili, et al. · arXiv (Cornell University)

Verify Smarter, Evolve Further: Efficient Harness Evolution through Behavior-Aware Verification

2026 · Xu, Jinghan, Zhang, Yikai, et al. · arXiv (Cornell University)

Selection-Invariant Communication Compilers for Privacy-Aware Multi-Agent LLM Workflows

2026 · Xu, Jinghan, Fan, Longze, et al. · arXiv (Cornell University)

MNC: Scope-Bound Semantic Declassification for Private LLM-Agent Communication

2026 · Xu, Jinghan, Fan, Longze, et al. · arXiv (Cornell University)

Policy-Backed Selective Regeneration under Tainted Inter-Agent Communication

2026 · Xu, Jinghan, Fan, Longze, et al. · arXiv (Cornell University)

Works count: 8

Citation count: 6

H-index: 2

I10-index: