影响力指数

30.21/100

前 24.2%

全站排名 #15,584

发表论文11 篇

平均评分4.3

年均产出5.5 篇/年

Yuepeng Hu

PhD student@Duke University·美国·OpenReview

研究方向

watermarking · diffusion model

4.7

Watermark-based Attribution of AI-Generated Content

ICLR 2026Poster

三作

4.7

Forging Image Watermarks by Reversing Watermark Removal Attacks

ICLR 2026Rejected

三作

4.5

Log-To-Leak: Prompt Injection Attacks on Tool-Using LLM Agents via Model Context Protocol

ICLR 2026Rejected

一作

3.5

Fingerprinting LLMs via Prompt Injection

ICLR 2026Withdrawn

一作

3.5

Attacking Logic with Logic: Reasoning Injection Attack to Large Reasoning Models

ICLR 2026Rejected

二作

3.0

Automatically Generating Visual Hallucination Test Cases for Multimodal Large Language Models

ICLR 2025Withdrawn

三作

3.5

SafeText: Safe Text-to-image Models via Aligning the Text Encoder

合作者 (16)

Yuepeng Hu

Watermark-based Attribution of AI-Generated Content

Forging Image Watermarks by Reversing Watermark Removal Attacks

Log-To-Leak: Prompt Injection Attacks on Tool-Using LLM Agents via Model Context Protocol

Fingerprinting LLMs via Prompt Injection

Attacking Logic with Logic: Reasoning Injection Attack to Large Reasoning Models

SafeText: Safe Text-to-image Models via Aligning the Text Encoder

A Transfer Attack to Image Watermarks

Stable Signature is Unstable: Removing Image Watermark from Diffusion Models

Watermark-based Detection and Attribution of AI-Generated Image

Automatically Generating Visual Hallucination Test Cases for Multimodal Large Language Models

SafeText: Safe Text-to-image Models via Aligning the Text Encoder