AI safety researcher

Ruoxi Jia

Virginia Tech

Google Scholar dblp OpenReview ORCID Homepage

Last 12 months8 papers

Evaluation and benchmarks 2Other 1

AI safety papers from the last year

On this site: 57 papers

October 2026

Selecting Long-Horizon Trajectories for Reliable and Efficient Terminal-Agent Training
Oct 5OtherarXiv ↗

June 2026

Beyond Similarity: Trustworthy Memory Search for Personal AI Agents
Jun 4arXiv ↗

May 2026

Memory-Induced Tool-Drift in LLM Agents
May 24Evaluation and benchmarksarXiv ↗
Remembering More, Risking More: Longitudinal Safety Risks in Memory-Equipped LLM Agents
May 18arXiv ↗
Mitigating Many-shot Jailbreak Attacks with One Single Demonstration
May 8arXiv ↗

March 2026

From Weak Cues to Real Identities: Evaluating Inference-Driven De-Anonymization in LLM Agents
Mar 19Evaluation and benchmarksarXiv ↗

January 2026

Authorship follows dblp and the author's own Google Scholar profile; only AI safety papers on this site are listed.