AI safety researcher

Yinpeng Dong

Google Scholar dblp

Last 12 months12 papers

Attacks 2Defenses 1Evaluation and benchmarks 1

AI safety papers from the last year

On this site: 91 papers

October 2026

Why Static Probes Fail: Context Determines Deception Feature Composition
Oct 20DefensesarXiv ↗

September 2026

Before the Script, Set the Stage: How Worldview Simulation Amplifies Psychologically Grounded Persuasion in Multi-Turn Jailbreaking
Sep 2AttacksarXiv ↗
Membership Inference in Fine-tuned Diffusion Language Models via Token-level Memorization Asymmetry
Sep 1AttacksarXiv ↗

August 2026

MasDrift: Benchmarking Authorization Preservation Across Multi-Agent Architectures
Aug 2Evaluation and benchmarksarXiv ↗

July 2026

May 2026

March 2026

February 2026

MemPot: Defending Against Memory Extraction Attack with Optimized Honeypots
Feb 7arXiv ↗

November 2025

AI Deception: Risks, Dynamics, and Controls
Nov 27arXiv ↗

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