AI safety researcher

Kui Ren

Zhejiang University

Google Scholar dblp ORCID Homepage

Last 12 months28 papers

Defenses 6Attacks 6Evaluation and benchmarks 3Other 1

AI safety papers from the last year

On this site: 97 papers

October 2026

Rubric Judgments Lack Uncertainty: Conformal Weighting for LLM Evaluation
Oct 15Evaluation and benchmarksarXiv ↗
Mitigating Cross-Modal Hallucinations by Correcting Modality Misrouting in Omnimodal LLMs
Oct 13DefensesarXiv ↗
ASCENT: First-Order Optimal Fine-Tuning with Recalibration for Safety--Utility Co-Enhancement
Oct 6DefensesarXiv ↗
Task-Aware Joint Pruning and Distillation for Efficient Audio Deepfake Detection
Oct 4OtherarXiv ↗

September 2026

CollageAttack: Exploiting Cross-Modal Alignment Flaws in T2I Models through Spatial Text Composition
Sep 29AttacksarXiv ↗
DeShortcut-Align: Decoupling Spurious Shortcuts for Robust Safety Alignment in Large Reasoning Models
Sep 28DefensesarXiv ↗
EOPSA: Efficient On-Policy Self-Distilled Safety Alignment
Sep 28DefensesarXiv ↗
Beyond Final Decisions: A Process-Centric Benchmark for Transparent AI-Assisted Peer Review
Sep 5Evaluation and benchmarksarXiv ↗

August 2026

ActBench: Self-Evolving Benchmark of Behavioral Safety in Cowork Agents
Aug 10Evaluation and benchmarksarXiv ↗

July 2026

May 2026

April 2026

March 2026

January 2026

A Causal Perspective for Enhancing Jailbreak Attack and Defense
Jan 31AttacksarXiv ↗
RerouteGuard: Understanding and Mitigating Adversarial Risks for LLM Routing
Jan 29DefensesarXiv ↗
SpatialJB: How Text Distribution Art Becomes the "Jailbreak Key" for LLM Guardrails
Jan 14AttacksarXiv ↗

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