Xinyang Han
5 papers · 2024–2025 · 3 venues · across top CS/AI venues
Authors are matched by name, so researchers who share a name ("Yang Liu", "Wei Wang") appear as a single profile and their paper counts, venues and achievements are combined. How identity is handled.
Achievements
Jump to papers ↓
🌍
Conference Polyglot · Bronze
🌉
Interdisciplinary Bridge · Bronze
🗺️
Taxonomy Completionist
(19)
Top co-authors
Research topics
Artificial Intelligence
(3)
Machine Learning
(3)
Core AI
(3)
Learning Types · Machine Learning
(3)
Computer Vision
(2)
Deep Learning
(2)
Natural Language Processing
(2)
Analysis
(2)
Learning Types · Deep Learning
(2)
Generation · Computer Vision
(1)
Generation · Natural Language Processing
(1)
Optimization & Theory
(1)
Processing
(1)
Resources & Methods
(1)
Keywords
large language model
(3)
safety alignment
(2)
semantic segmentation
(1)
image segmentation
(1)
adversarial learning
(1)
logistic regression
(1)
preference learning
(1)
self-supervised learning
(1)
dialogue generation
(1)
instance segmentation
(1)
model safety
(1)
ai safety
(1)
monte carlo tree search
(1)
preference modeling
(1)
diffusion model
(1)
adversarial defense
(1)
novel view synthesis
(1)
jailbreak attack
(1)
panoptic segmentation
(1)
jailbreak defense
(1)
Papers
AdaSteer: Your Aligned LLM is Inherently an Adaptive Jailbreak Defender
EMNLP 2025
Beware of Your Po! Measuring and Mitigating AI Safety Risks in Role-Play Fine-Tuning of LLMs
ACL 2025
Chain of Strategy Optimization Makes Large Language Models Better Emotional Supporter
EMNLP 2025
The More You See in 2D the More You Perceive in 3D
CVPR 2024
Unsupervised Universal Image Segmentation
CVPR 2024