Portfolio item number 1
Short description of portfolio item number 1
Posts by Collection
portfolio
Portfolio item number 2
Short description of portfolio item number 2
publications
Multi-Task Learning Improves Synthetic Speech Detection
Published in ICASSP 2022, 2022
This paper studies synthetic speech detection through the lens of multi-task learning.
Recommended citation: Yichuan Mo, and Shilin Wang. (2022). "Multi-Task Learning Improves Synthetic Speech Detection." ICASSP 2022.
DICE: Domain-attack Invariant Causal Learning for Improved Data Privacy Protection and Adversarial Robustness
Published in SIGKDD 2022, 2022
This paper presents a unified causal learning framework for data privacy protection and adversarial robustness.
Recommended citation: Qibing Ren, Yiting Chen, Yichuan Mo, Qitian Wu, and Junchi Yan. (2022). "DICE: Domain-attack Invariant Causal Learning for Improved Data Privacy Protection and Adversarial Robustness." SIGKDD 2022.
Improving Generative Adversarial Networks via Adversarial Learning in Latent Space
Published in NeurIPS 2022 (Spotlight, Top 5%), 2022
Spotlight paper at NeurIPS 2022.
Recommended citation: Yang Li, Yichuan Mo, Liangliang Shi, Junchi Yan, Xiaolu Zhang, and Jun Zhou. (2022). "Improving Generative Adversarial Networks via Adversarial Learning in Latent Space." NeurIPS 2022.
When Adversarial Training Meets Vision Transformers: Recipes from Training to Architecture
Published in NeurIPS 2022 (Spotlight, Top 5%) (First work to improve adversarial robustness of ViTs), 2022
Spotlight paper at NeurIPS 2022. Professor John Hopcroft described this as the first work to improve adversarial robustness for vision transformers.
Recommended citation: Yichuan Mo, Dongxian Wu, Yifei Wang, Yiwen Guo, and Yisen Wang. (2022). "When Adversarial Training Meets Vision Transformers: Recipes from Training to Architecture." NeurIPS 2022.
PID: Prompt-Independent Data Protection Against Latent Diffusion Models
Published in ICML 2024, 2024
This paper proposes prompt-independent data protection against latent diffusion models.
Recommended citation: Ang Li, Yichuan Mo, Mingjie Li, and Yisen Wang. (2024). "PID: Prompt-Independent Data Protection Against Latent Diffusion Models." ICML 2024.
TERD: A Unified Framework for Safeguarding Diffusion Models Against Backdoors
Published in ICML 2024 (First backdoor input detection method for diffusion models), 2024
This paper introduces a unified framework for detecting and mitigating backdoors in diffusion models.
Recommended citation: Yichuan Mo, Hui Huang, Mingjie Li, Ang Li, and Yisen Wang. (2024). "TERD: A Unified Framework for Safeguarding Diffusion Models Against Backdoors." ICML 2024.
Fight Back Against Jailbreaking via Prompt Adversarial Tuning
Published in NeurIPS 2024, 2024
This paper proposes prompt adversarial tuning as a defense against jailbreak attacks.
Recommended citation: Yichuan Mo, Yuji Wang, Zeming Wei, and Yisen Wang. (2024). "Fight Back Against Jailbreaking via Prompt Adversarial Tuning." NeurIPS 2024.
Are Smarter LLMs Safer? Exploring Safety-Reasoning Trade-offs in Prompting and Fine-Tuning
Published in arXiv 2025 (First to reveal the safety–reasoning capability trade-off), 2025
This preprint explores safety-reasoning trade-offs in prompting and fine-tuning of large language models.
Recommended citation: Ang Li, Yichuan Mo, Mingjie Li, Yifei Wang, and Yisen Wang. (2025). "Are Smarter LLMs Safer? Exploring Safety-Reasoning Trade-offs in Prompting and Fine-Tuning." arXiv preprint arXiv:2502.09673.
Generalist++: A Meta-learning Framework for Mitigating Trade-off in Adversarial Training
Published in arXiv 2025, 2025
This preprint proposes a meta-learning framework for mitigating trade-offs in adversarial training.
Recommended citation: Yisen Wang, Yichuan Mo, Hongjun Wang, Junyi Li, and Zhouchen Lin. (2025). "Generalist++: A Meta-learning Framework for Mitigating Trade-off in Adversarial Training." arXiv preprint arXiv:2510.13361.
On the Adversarial Transferability of Generalized “Skip Connections”
Published in TPAMI 2026 (Journal extension of SGM, original paper cited 400+ times on Google Scholar), 2026
This journal article studies adversarial transferability through the lens of generalized skip connections.
Recommended citation: Yisen Wang, Yichuan Mo, Dongxian Wu, Mingjie Li, Xingjun Ma, and Zhouchen Lin. (2026). "On the Adversarial Transferability of Generalized "Skip Connections"." TPAMI 2026.
Jailbreak and Guard Aligned Language Models with Only Few In-Context Demonstrations
Published in TPAMI 2026 (Adopted at scale by Anthropic), 2026
This journal article was later followed by Anthropic.
Recommended citation: Zeming Wei, Yifei Wang, Li Ang, Yichuan Mo, and Yisen Wang. (2026). "Jailbreak and Guard Aligned Language Models with Only Few In-Context Demonstrations." TPAMI 2026.
Decoding Large Language Diffusion Models with Foreseeing Movement
Published in ICLR 2026 DeLTa Workshop, 2026
This workshop paper studies how to decode large language diffusion models with foreseeing movement.
Recommended citation: Yichuan Mo, Quan Chen, Mingjie Li, Zeming Wei, and Yisen Wang. (2026). "Decoding Large Language Diffusion Models with Foreseeing Movement." ICLR 2026 DeLTa Workshop.
TrustLDM: Benchmarking Trustworthiness in Language Diffusion Model
Published in ICLR 2026 Trustworthy Workshop (First benchmark for evaluating trustworthiness of language diffusion models), 2026
This workshop paper benchmarks trustworthiness in language diffusion model.
Recommended citation: Yichuan Mo, Yukun Jiang, Yanbo Shi, Mingjie Li, Michael Backes, Yang Zhang, and Yisen Wang. (2026). "TrustLDM: Benchmarking Trustworthiness in Language Diffusion Model." ICLR 2026 Trustworthy Workshop.
SelfCAD: Protecting Your Efficient Reasoning Capabilities via Self-Cautious Insertion
Published in Preprint 2026, 2026
This preprint proposes a self-cautious insertion method to protect efficient reasoning capabilities of language models.
Recommended citation: Taiye Chen, Mingjie Li, Yichuan Mo, Shuo Feng, and Yisen Wang. (2026). "SelfCAD: Protecting Your Efficient Reasoning Capabilities via Self-Cautious Insertion."
talks
Talk 1 on Relevant Topic in Your Field
Published:
This is a description of your talk, which is a markdown file that can be all markdown-ified like any other post. Yay markdown!
Conference Proceeding talk 3 on Relevant Topic in Your Field
Published:
This is a description of your conference proceedings talk, note the different field in type. You can put anything in this field.
teaching
Teaching experience 1
Undergraduate course, University 1, Department, 2014
This is a description of a teaching experience. You can use markdown like any other post.
Teaching experience 2
Workshop, University 1, Department, 2015
This is a description of a teaching experience. You can use markdown like any other post.