|
Hi there, thanks for visiting my website! I am a junior undergraduate student majoring in Software Engineering at Xi'an Jiaotong University. I am also a member of the Baidu Big Data and AI Elite Class at Xi’an Jiaotong University. I am currently interning at the LUD Undergraduate Research Group at Xi'an Jiaotong University, under the supervision of Professor Minnan Luo. LUD is founded by Professor Minnan Luo and a group of undergraduates interested in NLP. During my time with the LUD, I have found many like-minded friends, and together we grow and collaborate. To date, dozens of students have published research papers as first authors during their undergraduate studies. My primary research interest lies in building a healthier and more transparent internet information ecosystem in the age of artificial intelligence. Recently, I have been focusing on the safety challenges posed by LLMs, including toxicity and bias. Additionally, I have been exploring privacy risks such as membership inference attacks(MIA). I am looking for master's or doctoral programs that can accept me in 2026! |
|
|
|
|
[2025-05] Our paper "DHMD: Uncertainty-based Dynamic LVLMs Integration for Efficient Harmful Meme Detection" has been accepted at ACL 2025@workshop NLP4PosImpact (non-archival)!
|
|
|
[2025-1-21] 🎉🎉🎉 Our paper "From Predictions to Analyses: Rationale-Augmented Fake News Detection with Large Vision-Language Models" has been accepted at
WWW 2025 Main Conference
(oral) !!! The Oral Presentation Schedule for conference can be found at this Link. We welcome everyone to connect and engage with us. |
|
|
[2024-11] Our paper "Unveiling Fake News with Adversarial Arguments Generated by Multimodal Large Language Models" has been accepted at COLING 2025 Main Conference (short paper)!
|
|
|
[2024-09] Our paper "From Predictions to Analyses: Explainable Rationales-Augmented Fake News Detection with Large Vision-Language Models" has been accepted at EMNLP 2024@workshop NLP for Positive Impact (non-archival)!
|
|
|
|
Submissions under review |
|
|
Submissions under review |
|
|
Submissions under review |
|
|
|
Xiaofan Zheng*, Xinghao Wang*, Minnan Luo ACL 2025@workshop NLP4PosImpact (non-archival) |
|
|
Xiaofan Zheng, Zinan Zeng, Heng Wang, Yuyang Bai, Yuhan Liu, Minnan Luo THE WEB CONFERENCE 2025, Web4Good track as oral presentation. link/ code/ oral presentation schedule Despite their remarkable reasoning skills, Large Vision-Language Models (LVLMs) struggle to reliably detect multimodal fake news, often falling short of specialized detection models. While LVLMs excel at dissecting news content from multiple angles, they stumble when synthesizing these insights into accurate judgments. Our proposed EARAM framework bridges this gap by adaptively extracting critical rationales from their analyses and generating trustworthy explanations. |
|
|
Xiaofan Zheng, Minnan Luo, Xinghao Wang COLING 2025, NLP Applications track (short paper) link/ code In the age of social media, detecting fake news, especially in multimodal content, is more critical than ever. Conventional methods often fall short due to their black-box nature and lack of real-world understanding. We leverage Multimodal Large Language Models (MLLMs) to introduce adversarial reasoning through debates. By generating opposing arguments and learning from contradictory evidence, our approach transforms the model into a skilled debate referee, achieving more effective multimodal fusion and fine-tuning. Extensive experiments on four datasets demonstrate that our method outperforms state-of-the-art techniques, providing a powerful new tool to combat fake news. |
|
|
Xiaofan Zheng, Zinan Zeng, Heng Wang, Yuyang Bai, Yuhan Liu, Minnan Luo EMNLP 2024@workshop NLP for Positive Impact (non-archival) link / poster/ code |
|
|
 |
Xi'an Jiaotong University
2022.09 - 2026.07 (Expected) B.E. in Software Engineering GPA: 87.2 / 100.0 Advisor: Prof. Minnan Luo |
|
|
 |
Luo lab Undergraduate Division (LUD) @ XJTU
Advisor: Prof. Minnan Luo |
|
|
|
Template courtesy: Jon Barron. |