刘玲｜南京大学｜学者主页

2026

Vision Verification Enhanced Fusion of VLMs for Efficient Visual Reasoning

S. Tekin, Yi-Chang Xu, Gao-Wen Liu, R. Kompella, Margaret L. Loper, Ling Liu

arXiv.org

DOI

2026

A Multi-Agent Perception-Action Alliance for Efficient Long Video Reasoning

Yi-Chang Xu, Gao-Wen Liu, R. Kompella, Tian-Sheng Huang, Si-Hao Hu, Fatih Ilhan, S. Tekin, Zachary Yahn, Ling Liu

arXiv.org

DOI

2026

MELT: A Behavioral Trace Dataset for High-Risk Memecoin Launch Detection

Si-Hao Hu, S. Tekin, Yi-Chang Xu, Ling Liu

2026

H3Fusion: Helpful, Harmless, Honest Fusion of Aligned LLMs

S. Tekin, Fatih Ilhan, Si-Hao Hu, Tian-Sheng Huang, Yi-Chang Xu, Zachary Yahn, Ling Liu

Conference of the European Chapter of the Association for Computational Linguistics

DOI

2026

Attention-aware Inference Optimizations for Large Vision-Language Models with Memory-efficient Decoding

Fatih Ilhan, Gao-Wen Liu, R. Kompella, S. Tekin, Tian-Sheng Huang, Zachary Yahn, Yi-Chang Xu, Ling Liu

arXiv.org

DOI

2025

Large Language Model based Smart Contract Auditing with LLMBugScanner

Yining Yuan, Yifei Wang, Yi-Chang Xu, Zachary Yahn, Si-Hao Hu, Ling Liu

arXiv.org

DOI

2025

Rebellion: Noise-Robust Reasoning Training for Audio Reasoning Models

Tian-Sheng Huang, Virat Shejwalkar, Oscar Chang, Milad Nasr, Ling Liu

arXiv.org

DOI

2025

Safety Tax: Safety Alignment Makes Your Large Reasoning Models Less Reasonable

Tian-Sheng Huang, Si-Hao Hu, Fatih Ilhan, S. Tekin, Zachary Yahn, Yi-Chang Xu, Ling Liu

arXiv.org

DOI

2025

FedHFT: Efficient Federated Fine-tuning with Heterogeneous Edge Clients

Fatih Ilhan, S. Tekin, Tian-Sheng Huang, Gao-Wen Liu, R. Kompella, Greg Eisenhauer, Y. Lin, C. Pu, Ling Liu

International Conference on Cognitive Machine Intelligence

DOI

2025

Adversarial Attention Perturbations for Large Object Detection Transformers

Zachary Yahn, S. Tekin, Fatih Ilhan, Si-Hao Hu, Tian-Sheng Huang, Yi-Chang Xu, Margaret L Loper, Ling Liu

IEEE International Conference on Computer Vision

DOI

2025

A Neurosymbolic Agent System for Compositional Visual Reasoning

Yi-Chang Xu, Gaowen Liu, R. Kompella, Si-Hao Hu, Fatih Ilhan, S. Tekin, Zachary Yahn, Ling Liu

2025

Virus: Harmful Fine-tuning Attack for Large Language Models Bypassing Guardrail Moderation

Tian-Sheng Huang, Si-Hao Hu, Fatih Ilhan, S. Tekin, Ling Liu

arXiv.org

DOI

2025

Multi-Agent Reinforcement Learning with Focal Diversity Optimization

S. Tekin, Fatih Ilhan, Tian-Sheng Huang, Si-Hao Hu, Zachary Yahn, Ling Liu

arXiv.org

DOI

2024

Robust Few-Shot Ensemble Learning with Focal Diversity-Based Pruning

S. Tekin, Fatih Ilhan, Tian-Sheng Huang, Si-Hao Hu, Ka-Ho Chow, M. Loper, Ling Liu

ACM Transactions on Intelligent Systems and Technology

DOI

2024

Harmful Fine-tuning Attacks and Defenses for Large Language Models: A Survey

Tian-Sheng Huang, Si-Hao Hu, Fatih Ilhan, S. Tekin, Ling Liu

ACM Computing Surveys

DOI

2024

PokeLLMon: A Human-Parity Agent for Pokemon Battles with Large Language Models

Si-Hao Hu, Tian-Sheng Huang, Ling Liu

arXiv.org

DOI

2024

Personalized Privacy Protection Mask Against Unauthorized Facial Recognition

Ka-Ho Chow, Si-Hao Hu, Tian-Sheng Huang, Ling Liu

European Conference on Computer Vision

DOI

2024

Resource- Efficient Transformer Pruning for Finetuning of Large Models

Fatih Ilhan, Gong Su, S. Tekin, Tian-Sheng Huang, Si-Hao Hu, Ling Liu

Computer Vision and Pattern Recognition

DOI

2024

LLM-TOPLA: Efficient LLM Ensemble by Maximising Diversity

S. Tekin, Fatih Ilhan, Tian-Sheng Huang, Si-Hao Hu, Ling Liu

Conference on Empirical Methods in Natural Language Processing

DOI

2024

Vaccine: Perturbation-aware Alignment for Large Language Model

Tian-Sheng Huang, Si-Hao Hu, Ling Liu

Neural Information Processing Systems

DOI

2024

Antidote: Post-fine-tuning Safety Alignment for Large Language Models against Harmful Fine-tuning

Tian-Sheng Huang, Gautam Bhattacharya, Pratik Joshi, J. Kimball, Ling Liu

International Conference on Machine Learning

DOI

2024

A Survey on Large Language Model-Based Game Agents

Si-Hao Hu, Tian-Sheng Huang, Fatih Ilhan, S. Tekin, Gao-Wen Liu, R. Kompella, Ling Liu

arXiv.org

DOI

2024

Booster: Tackling Harmful Fine-tuning for Large Language Models via Attenuating Harmful Perturbation

Tian-Sheng Huang, Si-Hao Hu, Fatih Ilhan, S. Tekin, Ling Liu

International Conference on Learning Representations

DOI

2024

Diversity-driven Privacy Protection Masks Against Unauthorized Face Recognition

Ka-Ho Chow, Si-Hao Hu, Tian-Sheng Huang, Fatih Ilhan, Wenqi Wei, Ling Liu

Proceedings on Privacy Enhancing Technologies

DOI

2024

Lisa: Lazy Safety Alignment for Large Language Models against Harmful Fine-tuning Attack

Tian-Sheng Huang, Si-Hao Hu, Fatih Ilhan, S. Tekin, Ling Liu

Neural Information Processing Systems

DOI

2023

Large Language Model-Powered Smart Contract Vulnerability Detection: New Perspectives

Si-Hao Hu, Tian-Sheng Huang, Fatih Ilhan, S. Tekin, Ling Liu

International Conference on Trust, Privacy and Security in Intelligent Systems and Applications

DOI

2023

Scalable Federated Learning with System Heterogeneity

Fatih Ilhan, Gong Su, Qingyan Wang, Ling Liu

IEEE International Conference on Distributed Computing Systems

DOI

2023

Lockdown: Backdoor Defense for Federated Learning with Isolated Subspace Training

Tian-Sheng Huang, Si-Hao Hu, Ka-Ho Chow, Fatih Ilhan, S. Tekin, Ling Liu

Neural Information Processing Systems

DOI

刘玲

精选作品

期刊论文

Vision Verification Enhanced Fusion of VLMs for Efficient Visual Reasoning

A Multi-Agent Perception-Action Alliance for Efficient Long Video Reasoning

MELT: A Behavioral Trace Dataset for High-Risk Memecoin Launch Detection

H3Fusion: Helpful, Harmless, Honest Fusion of Aligned LLMs

Attention-aware Inference Optimizations for Large Vision-Language Models with Memory-efficient Decoding

Large Language Model based Smart Contract Auditing with LLMBugScanner

Rebellion: Noise-Robust Reasoning Training for Audio Reasoning Models

Safety Tax: Safety Alignment Makes Your Large Reasoning Models Less Reasonable

FedHFT: Efficient Federated Fine-tuning with Heterogeneous Edge Clients

Adversarial Attention Perturbations for Large Object Detection Transformers

A Neurosymbolic Agent System for Compositional Visual Reasoning

Virus: Harmful Fine-tuning Attack for Large Language Models Bypassing Guardrail Moderation

Multi-Agent Reinforcement Learning with Focal Diversity Optimization

Robust Few-Shot Ensemble Learning with Focal Diversity-Based Pruning

Harmful Fine-tuning Attacks and Defenses for Large Language Models: A Survey

PokeLLMon: A Human-Parity Agent for Pokemon Battles with Large Language Models

Personalized Privacy Protection Mask Against Unauthorized Facial Recognition

Resource- Efficient Transformer Pruning for Finetuning of Large Models

LLM-TOPLA: Efficient LLM Ensemble by Maximising Diversity

Vaccine: Perturbation-aware Alignment for Large Language Model

Antidote: Post-fine-tuning Safety Alignment for Large Language Models against Harmful Fine-tuning

A Survey on Large Language Model-Based Game Agents

Booster: Tackling Harmful Fine-tuning for Large Language Models via Attenuating Harmful Perturbation

Diversity-driven Privacy Protection Masks Against Unauthorized Face Recognition

Lisa: Lazy Safety Alignment for Large Language Models against Harmful Fine-tuning Attack

Large Language Model-Powered Smart Contract Vulnerability Detection: New Perspectives

Scalable Federated Learning with System Heterogeneity

Lockdown: Backdoor Defense for Federated Learning with Isolated Subspace Training