影响力指数

62.5/100

前 4.1%

全站排名 #2,667

发表论文10 篇

平均评分5.2

年均产出5.0 篇/年

Albert No

Associate Professor@Yonsei University·韩国·OpenReview

研究方向

Diffusion Generative Models · Differential Privacy · Information Theory · Machine Unlearning · LLM Quantization · LLM Safety

6.0

Rainbow Padding: Mitigating Early Termination in Instruction-Tuned Diffusion LLMs

ICLR 2026Poster

通讯

5.5

A2D: Any-Order, Any-Step Safety Alignment for Diffusion Language Models

ICLR 2026Poster

通讯

5.3

Rethinking Benign Relearning: Syntax as the Hidden Driver of Unlearning Failures

ICLR 2026Poster

通讯

3.5

Breaking the Reversal Curse: How Masked Diffusion Models Achieve Reverse Inference

ICLR 2026Rejected

通讯

3.3

dMARK: Decoding-Guided Watermarking for Discrete Diffusion Language Models

ICLR 2026Rejected

二作

3.3

DUSK: Do Not Unlearn Shared Knowledge

ICLR 2026Withdrawn

通讯

6.6

Understanding and Mitigating Memorization in Generative Models via Sharpness of Probability Landscapes

合作者 (17)

Albert No

Rainbow Padding: Mitigating Early Termination in Instruction-Tuned Diffusion LLMs

A2D: Any-Order, Any-Step Safety Alignment for Diffusion Language Models

Rethinking Benign Relearning: Syntax as the Hidden Driver of Unlearning Failures

Breaking the Reversal Curse: How Masked Diffusion Models Achieve Reverse Inference

dMARK: Decoding-Guided Watermarking for Discrete Diffusion Language Models

DUSK: Do Not Unlearn Shared Knowledge

Understanding and Mitigating Memorization in Generative Models via Sharpness of Probability Landscapes

SAFEPATH: Preventing Harmful Reasoning in Chain-of-Thought via Early Alignment

Information-Theoretic Discrete Diffusion

AN INFORMATION THEORETIC EVALUATION METRIC FOR STRONG UNLEARNING