Publications

2026

  1. Internal Safety Collapse in Frontier Large Language Models [Code]

    Yutao Wu, Xiao Liu, Hanxun Huang, Yige Li, Xiang Zheng, Yifeng Gao, Cong Wang, Bo Li, Xingjun Ma, Yu-Gang Jiang. NeurIPS, Sydney, Australia, 2026.

  2. VEX-Bench: Benchmarking Verification Complexity of LLM-Generated Misinformation [Code]

    Hanxun Huang, Yutao Wu, Qizhou Wang, Silvia Montaña-Niño, Yige Li, Xiang Zheng, Elif Buse Doyuran, Phoebe Matich, Xiao Liu, Xingjun Ma, Sarah Erfani, Christopher Leckie. NeurIPS, Sydney, Australia, 2026.

  3. ShadowFPT: Backdooring Federated Prompt Tuning via Shadow Triggers

    Kun Zhai, Teng Li, Yunhao Feng, Xingjun Ma. NeurIPS, Sydney, Australia, 2026.

  4. Towards Multi-Human-Value Alignment via Value Localization in LLMs

    Xueqi Ma, Yanbei Jiang, Xingjun Ma, James Bailey, Sarah Erfani. NeurIPS, Sydney, Australia, 2026.

  5. AgentHazard: A Benchmark for Evaluating Harmful Behavior in Computer-Use Agents [Code]

    Yifan Ding, Yunhao Feng, Yifeng Gao, Yige Li, Yutao Wu, Yanming Guo, Yingshui Tan, Kun Zhai, Xingjun Ma. MM, Rio de Janeiro, Brazil, 2026.

  6. RedDiffuser: Auditing Multimodal Safety Failures in Vision-Language Models via Reinforced Diffusion

    Ruofan Wang, Xingjun Ma. MM, Rio de Janeiro, Brazil, 2026.

  7. BackdoorVLM: A Benchmark for Backdoor Attacks and Defenses on Vision-Language Models [Code]

    Juncheng Li, Yige Li, Hanxun Huang, Yunhao Chen, Xin Wang, Yixu Wang, Xingjun Ma, Yu-Gang Jiang. MM, Rio de Janeiro, Brazil, 2026.

  8. Agent4POI: Agentic context-conditioned affordance reasoning for Multimodal Point-of-Interest Recommendation

    Jinze Wang, Yangchen Zeng, Tiehua Zhang, Lu Zhang, Yuze Liu, Yongchao Liu, Xingjun Ma, Zhu Sun. MM, Rio de Janeiro, Brazil, 2026.

  9. DropVLA: An Action-Level Backdoor Attack on Vision-Language-Action Models

    Zonghuan Xu, Jiayu Li, Yunhan Zhao, Xiang Zheng, Xingjun Ma, Yu-Gang Jiang. IROS, Pittsburgh, Pennsylvania, USA, 2026.

  10. Perturbation Effects on Robustness and Individual Fairness

    Xuran Li, Hao Xue, Peng Wu, Xingjun Ma, Zhen Zhang, Huaming Chen, Flora D. Salim. KDD, Jeju Island, South Korea, 2026.

  11. FakeWorld 1.0: An Omni-modal Benchmark for Fake Media and Content

    Yifeng Gao, Yifan Ding, Li Wang, Feida Huang, Ye Sun, Yixu Wang, Xin Wang, Yutao Wu, Hanxun Huang, Yunhao Feng, Yingshui Tan, Xingjun Ma, Yu-Gang Jiang. ICML, Seoul, South Korea, 2026.

  12. Just Ask: Curious Code Agents Reveal System Prompts in Frontier LLMs [Code] [Project]

    Xiang Zheng, Yutao Wu, Hanxun Huang, Yige Li, Xingjun Ma, Bo Li, Yu-Gang Jiang, Cong Wang. ICML, Seoul, South Korea, 2026.

  13. SciAgentGym: Benchmarking Multi-Step Scientific Tool-Use in LLM Agents

    Yujiong Shen, Yajie Yang, Zhiheng Xi, Binze Hu, Huayu Sha, Qiyuan Peng, Jiazheng Zhang, Junlin Shang, Jixuan Huang, Yutao Fan, Jingqi Tong, Shihan Dou, Ming Zhang, LEI BAI, Zhenfei Yin, Tao Gui, Xingjun Ma, Qi Zhang, Xuanjing Huang, Yu-Gang Jiang. ICML, Seoul, South Korea, 2026.

  14. AudioMosaic: Contrastive Masked Audio Representation Learning

    Hanxun Huang, Qizhou Wang, Xingjun Ma, Cihang Xie, Christopher Leckie, Sarah Monazam Erfani. ICML, Seoul, South Korea, 2026.

  15. Towards Context-Invariant Safety Alignment for Large Language Models

    Yixu Wang, Yang Yao, Xin Wang, Yifeng Gao, Yan Teng, Xingjun Ma, Yingchun Wang. ICML, Seoul, South Korea, 2026.

  16. MESA: Improving MoE Safety Alignment via Decentralized Expertise

    Yitong Sun, Yao Huang, Teng Li, Ranjie Duan, Yichi Zhang, Xingjun Ma, Hui Xue, Xingxing Wei. ICML, Seoul, South Korea, 2026.

  17. RA-Det: Towards Universal Detection of AI-Generated Images via Robustness Asymmetry [Code]

    Xinchang Wang, Yunhao Chen, Yuechen Zhang, Congcong Bian, Zihao Guo, Xingjun Ma, Hui Li. ICML, Seoul, South Korea, 2026.

  18. Deliberative Searcher: Improving LLM Reliability via Reinforcement Learning with Constraints

    Zhenyun Yin, Shujie Wang, Xuhong Wang, Xingjun Ma, Yingchun Wang. ACL, San Diego, California, USA, 2026. [Main, Oral]

  19. BackdoorAgent: A Unified Framework for Backdoor Attacks on LLM-based Agents [Code]

    Yunhao Feng, Yige Li, Yutao Wu, Yingshui Tan, Yanming Guo, Yifan Ding, Kun Zhai, Xingjun Ma, Yu-Gang Jiang. ACL, San Diego, California, USA, 2026. [Findings]

  20. AgenticEval: Toward Agentic and Self-Evolving Safety Evaluation of Large Language Models

    Yixu Wang, Xin Wang, Yang Yao, Xinyuan Li, Xibang Yang, Yan Teng, Xingjun Ma, Yingchun Wang. ACL, San Diego, California, USA, 2026. [Findings]

  21. OmniLottie: Generating Vector Animations via Parameterized Lottie Tokens [Code] [Project Page] [Hugging Face]

    Yiying Yang, Wei Cheng, Sijin Chen, Honghao Fu, Xianfang Zeng, Yujun Cai, Gang YU, Xingjun Ma. CVPR, Denver CO, USA, 2026.

  22. GenBreak: Red Teaming Text-to-Image Generation Using Large Language Models

    Zilong Wang, Xiang Zheng, Xiaosen Wang, Bo Wang, Xingjun Ma. CVPR, Denver CO, USA, 2026.

  23. WithAnyone: Towards controllable and id consistent image generation [Code] [Project]

    Hengyuan Xu, Wei Cheng, Peng Xing, Yixiao Fang, Shuhan Wu, Rui Wang, Xianfang Zeng, Daxin Jiang, Gang Yu, Xingjun Ma, Yu-Gang Jiang. ICLR, Rio de Janeiro, Brazil, 2026.

  24. Toward Universal and Transferable Jailbreak Attacks on Vision-Language Models [Code]

    Kaiyuan Cui, Yige Li, Yutao Wu, Xingjun Ma, Sarah Erfani, Christopher Leckie, Hanxun Huang. ICLR, Rio de Janeiro, Brazil, 2026.

  25. SIDE: Surrogate Conditional Data Extraction from Diffusion Models

    Yunhao Chen, Shujie Wang, Difan Zou, Xingjun Ma. AAAI, Singapore, 2026.

  26. PaperAsk: A Benchmark for Reliability Evaluation of LLMs in Paper Search and Reading

    Yutao Wu, Xiao Liu, Yunhao Feng, Jiale Ding, Xingjun Ma. WWW, Dubai, United Arab Emirates, 2026.

  27. Coarse-to-Fine Open-Set Graph Node Classification with Large Language Models

    Xueqi Ma, Xingjun Ma, Sarah Erfani, Danilo Mandic, James Bailey. AAAI, Singapore, 2026.

  28. Do We Really Need SFT? Prompt-as-Policy over Knowledge Graphs for Cold-start Next POI Recommendation

    Jinze Wang, Lu Zhang, Tiehua Zhang, Yiyang Cui, Zhishu Shen, Yuze Liu, Xingjun Ma, Jiong Jin. CIKM, Rome, Italy, 2026.

  29. NAP-Tuning: Neural Augmented Prompt Tuning for Adversarially Robust Vision-Language Models

    Jiaming Zhang, Xin Wang, Xingjun Ma, Lingyu Qiu, Yu-Gang Jiang, Jitao Sang. IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI), 2026.

  30. FedEGG: Federated Learning with Explicit Global Guidance

    Kun Zhai, Yifeng Gao, Yunhao Feng, Wei Gao, Xingjun Ma, Yu-Gang Jiang. Frontiers of Computer Science (FCS), 2026.

  31. On the Adversarial Transferability of Generalized "Skip Connections" [Code]

    Yisen Wang, Yichuan Mo, Dongxian Wu, Mingjie Li, Xingjun Ma, Zhouchen Lin. IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI), 2026.

  32. OpenRedRL: A Light-Weight Benchmark for Reinforcement Fine-Tuning-Based Red Teaming [Code]

    Xiang Zheng, Xingjun Ma, Wei-Bin Lee, Cong Wang. Frontiers of Computer Science (FCS), 2026.

  33. Defense-to-attack: Bypassing weak defenses enables stronger jailbreaks in Vision-Language Models

    Yunhan Zhao, Xiang Zheng, Xingjun Ma. Pattern Recognition, 2026.

  34. Learnable Coreset Selection for Graph Active Learning

    Xueqi Ma, Xingjun Ma, Sarah Erfani, James Bailey. Transactions on Machine Learning Research (TMLR), 2026.

2025

  1. BackdoorLLM: A Comprehensive Benchmark for Backdoor Attacks and Defenses on Large Language Models [Code]

    Yige Li, Hanxun Huang, Yunhan Zhao, Xingjun Ma, Jun Sun. NeurIPS, San Diego, USA, 2025.

  2. SafeVid: Toward Safety Aligned Video Large Multimodal Models [Dataset]

    Yixu Wang, Jiaxin Song, Yifeng Gao, Xin Wang, Yang Yao, Yan Teng, Xingjun Ma, Yingchun Wang, Yu-Gang Jiang. NeurIPS, San Diego, USA, 2025.

  3. OmniSVG: A Unified Scalable Vector Graphics Generation Model [Code] [Project Page] [Hugging Face]

    Yiying Yang, Wei Cheng, Sijin Chen, Xianfang Zeng, Fukun Yin, Jiaxu Zhang, Liao Wang, Gang YU, Xingjun Ma, Yu-Gang Jiang. NeurIPS, San Diego, USA, 2025.

  4. SAMA: Towards Multi-Turn Referential Grounded Video Chat with Large Language Models

    Ye Sun, Hao Zhang, Henghui Ding, Tiehua Zhang, Xingjun Ma, Yu-Gang Jiang. NeurIPS, San Diego, USA, 2025.

  5. JailBound: Jailbreaking Internal Safety Boundaries of Vision-Language Models

    Jiaxin Song, Yixu Wang, Jie Li, Xuan Tong, Rui Yu, Yan Teng, Xingjun Ma, Yingchun Wang. NeurIPS, San Diego, USA, 2025.

  6. Safety at Scale: A Comprehensive Survey of Large Model and Agent Safety[Code]

    Xingjun Ma, Yifeng Gao, Yixu Wang, Ruofan Wang, Xin Wang, Ye Sun, Yifan Ding, Hengyuan Xu, Yunhao Chen, Yunhan Zhao, Hanxun Huang, Yige Li, Yutao Wu, Jiaming Zhang, Xiang Zheng, Yang Bai, Zuxuan Wu, Xipeng Qiu, Jingfeng Zhang, Yiming Li, Xudong Han, Haonan Li, Jun Sun, Cong Wang, Jindong Gu, Baoyuan Wu, Siheng Chen, Tianwei Zhang, Yang Liu, Mingming Gong, Tongliang Liu, Shirui Pan, Cihang Xie, Tianyu Pang, Yinpeng Dong, Ruoxi Jia, Yang Zhang, Shiqing Ma, Xiangyu Zhang, Neil Gong, Chaowei Xiao, Sarah Erfani, Tim Baldwin, Bo Li, Masashi Sugiyama, Dacheng Tao, James Bailey, Yu-Gang Jiang. Foundations and Trends® in Privacy and Security, 2025.

  7. Shortcuts Everywhere and Nowhere: Exploring Multi-Trigger Backdoor Attacks[Code]

    Yige Li, Jiabo He, Hanxun Huang, Jun Sun, Xingjun Ma, Yu-Gang Jiang. TDSC, 2025.

  8. BadPatch: Diffusion-Based Generation of Physical Adversarial Patches[Code] [AdvT-shirt-1K Dataset]

    Zhixiang Wang, Xingjun Ma, Yu-Gang Jiang. ICCV Workshop Findings, Honolulu, Hawai'i, 2025.

  9. IDEATOR: Jailbreaking and Benchmarking Large Vision-Language Models Using Themselves[Code]

    Ruofan Wang, Juncheng Li, Yixu Wang, Bo Wang, Xiaosen Wang, Yan Teng, Yingchun Wang, Xingjun Ma, Yu-Gang Jiang. ICCV, Honolulu, Hawai'i, 2025.

  10. Free-Form Motion Control: Controlling the 6D Poses of Camera and Objects in Video Generation

    Xincheng Shuai, Henghui Ding, Zhenyuan Qin, Hao Luo, Xingjun Ma, Dacheng Tao. ICCV, Honolulu, Hawai'i, 2025.

  11. StolenLoRA: Exploring LoRA Extraction Attacks via Synthetic Data

    Yixu Wang, Yan Teng, Yingchun Wang, Xingjun Ma. ICCV, Honolulu, Hawai'i, 2025.

  12. T2UE: Generating Unlearnable Examples from Text Descriptions

    Xingjun Ma, Hanxun Huang, Tianwei Song, Ye Sun, Yifeng Gao, Yu-Gang Jiang. MM, Dublin, Ireland, 2025.

  13. FedAPT: Federated Adversarial Prompt Tuning for Vision-Language Models

    Kun Zhai, Siheng Chen, Xingjun Ma, Yu-Gang Jiang. MM, Dublin, Ireland, 2025.

  14. From Failures to Fixes: LLM-Driven Scenario Repair for Self-Evolving Autonomous Driving[Code]

    Xinyu Xia, Xingjun Ma, Yunfeng hu, Qu Ting, Hong Chen, Xun Gong. MM, Dublin, Ireland, 2025.

  15. BrokenVideos: A Benchmark Dataset for Fine-Grained Artifact Localization in AI-Generated Videos[Code]

    Jiahao Lin, Weixuan Peng, Bojia Zi, Yifeng Gao, Xianbiao Qi, Xingjun Ma, Yu-Gang Jiang. Dataset Track, MM, Dublin, Ireland, 2025.

  16. X-Transfer Attacks: Towards Super Transferable Adversarial Attacks on CLIP[Code] [Project] [HuggingFace demo]

    Hanxun Huang, Sarah Erfani, Yige Li, Xingjun Ma, James Bailey. ICML, Vancouver, Canada, 2025.

  17. Towards Million-Scale Adversarial Robustness Evaluation With Stronger Individual Attacks[Project]

    Yong Xie, Weijie Zheng, Hanxun Huang, Guangnan Ye, Xingjun Ma. CVPR, Nashville TN, 2025.

  18. TAPT: Test-Time Adversarial Prompt Tuning for Robust Inference in Vision-Language Models[Code]

    Xin Wang, Kai Chen, Jiaming Zhang, Jingjing Chen, Xingjun Ma. CVPR, Nashville TN, 2025.

  19. AnyAttack: Towards Large-scale Self-supervised Adversarial Attacks on Vision-Language Models[Code] [Project]

    Jiaming Zhang, Junhong Ye, Xingjun Ma, Yige Li, Yunfan Yang, Yunhao Chen, Jitao Sang, Dit-Yan Yeung. CVPR, Nashville TN, 2025.

  20. BlueSuffix: Reinforced Blue Teaming for Vision-Language Models Against Jailbreak Attacks[Code]

    Yunhan Zhao, Xiang Zheng, Lin Luo, Yige Li, Xingjun Ma, Yu-Gang Jiang. ICLR, Singapore, 2025.

  21. Detecting Backdoor Samples in Contrastive Language Image Pretraining[Code] [Project]

    Hanxun Huang, Sarah Erfani, Yige Li, Xingjun Ma, James Bailey. ICLR, Singapore, 2025.

  22. AIM: Additional Image Guided Generation of Transferable Adversarial Attacks

    Teng Li, Xingjun Ma, Yu-Gang Jiang. AAAI, Philadelphia, USA, 2025.

  23. CALM: Curiosity-Driven Auditing for Large Language Models[Code]

    Xiang Zheng, Longxiang WANG, Yi Liu, Xingjun Ma, Chao Shen, Cong Wang. AAAI, Philadelphia, USA, 2025.

  24. HoneypotNet: Backdoor Attacks Against Model Extraction

    Yixu Wang, Tianle Gu, Yan Teng, Yingchun Wang, Xingjun Ma. AAAI, Philadelphia, USA, 2025.

  25. MMFair: Fair Learning via Min-Min Optimization

    Kejie Fang, Kun Zhai, Xingjun Ma. CIKM, Seoul, Korea, 2025.

  26. Optimizing Cross-Client Domain Coverage for Federated Instruction Tuning of Large Language Models

    Zezhou Wang, Yaxin Du, Xingjun Ma, Yu-Gang Jiang, Zhuzhong Qian, Siheng Chen. EMNLP 2025 Findings, Suzhou, China, 2025.

  27. Learning from Heterogeneity: A Dynamic Learning Framework for Hypergraphs [Code]

    Tiehua Zhang, Yuze Liu, Zhishu Shen, Xingjun Ma, Peng Qi, Zhijun Ding, Jiong Jin. TAI, To appear in 2025.

2024

  1. UnSeg: One Universal Unlearnable Example Generator is Enough against All Image Segmentation

    Ye Sun, Hao Zhang, Tiehua Zhang, Xingjun Ma, Yu-Gang Jiang. NeurIPS, Vancouver, Canada, 2024.

  2. ModelLock: Locking Your Model With a Spell

    Yifeng Gao, Yuhua Sun, Xingjun Ma, Zuxuan Wu, Yu-Gang Jiang. MM, Melbourne, Australia, 2024.

  3. White-box Multimodal Jailbreaks Against Large Vision-Language Models

    Ruofan Wang, Xingjun Ma, Hanxu Zhou, Chuanjun Ji, Guangnan Ye, Yu-Gang Jiang. MM, Melbourne, Australia, 2024.

  4. AdvQDet: Detecting Query-Based Adversarial Attacks with Adversarial Contrastive Prompt Tuning [Code]

    Xin Wang, Kai Chen, Xingjun Ma, Zhineng Chen, Jingjing Chen, Yu-Gang Jiang. MM, Melbourne, Australia, 2024.

  5. Fuse Your Latents: Video Editing with Multi-source Latent Diffusion Models [Code]

    Tianyi Lu, Xing Zhang, Jiaxi Gu, Hang Xu, Renjing Pei, Songcen Xu, Xingjun Ma, Zuxuan Wu. MM, Melbourne, Australia, 2024.

  6. Adversarial Prompt Tuning for Vision-Language Models [Code]

    Jiaming Zhang, Xingjun Ma, Xin Wang, Lingyu Qiu, Jiaqi Wang, Yu-Gang Jiang, Jitao Sang. ECCV, MiCo Milano, Italy, 2024.

  7. Constrained Intrinsic Motivation for Reinforcement Learning [Code]

    Xiang Zheng, Xingjun Ma, Chao Shen, Cong Wang. IJCAI, Jeju, Korea, 2024.

  8. Toward Evaluating Robustness of Reinforcement Learning with Adversarial Policy [Code]

    Xiang Zheng, Xingjun Ma, Shengjie Wang, Xinyu Wang, Chao Shen, Cong Wang. DSN, Brisbane, Australia, 2024.

  9. VeriFi: Towards Verifiable Federated Unlearning

    Xiangshan Gao, Xingjun Ma, Jingyi Wang, Youcheng Sun, Bo Li, Shouling Ji, Peng Cheng, Jiming Chen. TDSC, 2024. (Best Paper Runner-up)

  10. Fake Alignment: Are LLMs Really Aligned Well?

    Yixu Wang, Yan Teng, Kexin Huang, Chengqi Lyu, Songyang Zhang, Wenwei Zhang, Xingjun Ma, Yu-Gang Jiang, Yu Qiao, Yingchun Wang. NAACL, Mexico City, Mexico, 2024.

  11. LDReg: Local Dimensionality Regularized Self-Supervised Learning [Code]

    Hanxun Huang, Ricardo J. G. B. Campello, Sarah M. Erfani, Xingjun Ma, Michael E. Houle, James Bailey. ICLR, Vienna, Austria, 2024.

  12. Unlearnable Examples For Time Series

    Yujing Jiang, Xingjun Ma, Sarah Monazam Erfani and James Bailey. PAKDD, 2024.

2023

  1. Reconstructive Neuron Pruning for Backdoor Defense [Code]

    Yige Li, Xixiang Lyu, Xingjun Ma, Nodens Koren, Lingjuan Lyu, Bo Li, Yu-Gang Jiang. ICML, Hawaii, USA, 2023.

  2. Unlearnable Clusters: Towards Label-agnostic Unlearnable Examples [Code]

    Jiaming Zhang, Xingjun Ma, Qi Yi, Jitao Sang, Yu-Gang Jiang, Yaowei Wang, Changsheng Xu. CVPR, Vancouver, Canada, 2023.

  3. Distilling Cognitive Backdoor Patterns within an Image [Code]

    Hanxun Huang, Xingjun Ma, Sarah M. Erfani, James Bailey. ICLR, Kigali, Rwanda, 2023.

  4. Transferable Unlearnable Examples[Code]

    Jie Ren, Han Xu, Yuxuan Wan, Xingjun Ma, Lichao Sun, Jiliang Tang. ICLR, Kigali, Rwanda, 2023.

  5. On the Importance of Spatial Relations for Few-shot Action Recognition

    Yilun Zhang, Yuqian Fu, Xingjun Ma, Lizhe Qi, Jingjing Chen, Zuxuan Wu, Yu-Gang Jiang. MM, Ottawa, Canada, 2023.

  6. Backdoor Attacks on Time Series: A Generative Approach[Code]

    Yujing Jiang, Xingjun Ma, Sarah M. Erfani, James Bailey. SaTML, 2023.

  7. Relationships between tail entropies and local intrinsic dimensionality and their use for estimation and feature representation

    James Bailey, Michael E. Houle, Xingjun Ma. Information Systems (2023): 102245.

  8. Imbalanced Gradients: A Subtle Cause of Overestimated Adversarial Robustness[Code]

    Xingjun Ma*, Linxi Jiang*, Hanxun Huang, Zejia Weng, James Bailey, Yu-Gang Jiang. Machine Learning (2023): 1-26.

  9. Query-efficient Black-box Adversarial Attacks on Automatic Speech Recognition

    Chuxuan Tong, Xi Zheng, Jianhua Li, Xingjun Ma, Longxiang Gao, Yong Xiang. To appear in TASLP.

2022

  1. Local Intrinsic Dimensionality, Entropy and Statistical Divergences

    James Bailey, Michael E. Houle, Xingjun Ma. Entropy 24(9), 1220, 2022.

  2. Few-Shot Backdoor Attacks on Visual Object Tracking [Code]

    Yiming Li, Haoxiang Zhong, Xingjun Ma, Yong Jiang, Shu-Tao Xia. ICLR, 2022.

  3. CalFAT: Calibrated Federated Adversarial Training with Label Skewness

    Chen Chen, Yuchen Liu, Xingjun Ma, Lingjuan Lyu. NeurIPS, 2022.

  4. Copy, Right? A Testing Framework for Copyright Protection of Deep Learning Models [Code]

    Jialuo Chen, Jingyi Wang, Tinglan Peng, Youcheng Sun, Peng Cheng, Shouling Ji, Xingjun Ma, Bo Li, Dawn Song. Oakland, 2022.

  5. Backdoor Attacks on Crowd Counting [Code]

    Yuhua Sun, Tailai Zhang, Xingjun Ma, Pan Zhou, Jian Lou, Zichuan Xu, Xing Di, Yu Cheng, Lichao Sun. MM, 2022.

  6. Fine-mixing: Mitigating Backdoors in Fine-tuned Language Models

    Zhiyuan Zhang, Lingjuan Lyu, Xingjun Ma, Chenguang Wang, Xu Sun. EMNLP, 2022.

  7. Privacy and Robustness in Federated Learning: Attacks and Defenses

    Lingjuan Lyu, Han Yu, Xingjun Ma, Chen Chen, Lichao Sun, Jun Zhao, Qiang Yang, Philip Yu. TNNLS, 2022.

  8. QuoTe: Quality-oriented Testing for Deep Learning Systems

    Jialuo Chen, Jingyi Wang*, Xingjun Ma, Youcheng Sun, Jun Sun, Peixin Zhang and Peng Cheng. TOSEM (accepted in 2022).

  9. Machine learning guided alloy design of high-temperature NiTiHf shape memory alloys

    Udesh M.H.U. Kankanamge, Johannes Reiner, Xingjun Ma, Santiago Corujeira Gallo, Wei Xu. Journal of Materials Science. (2022 Robert W. Cahn Best Paper Award)

2021

  1. Exploring Architectural Ingredients of Adversarially Robust Deep Neural Networks[Code]

    Hanxun Huang, Yisen Wang, Sarah M. Erfani, Quanquan Gu, James Bailey, Xingjun Ma. NeurIPS, 2021.

  2. Alpha-IoU: A Family of Power Intersection over Union Losses for Bounding Box Regression[Code]

    Jiabo He, Sarah M. Erfani, Xingjun Ma, James Bailey, Ying Chi, Xian-Sheng Hua. NeurIPS, 2021.

  3. Anti-Backdoor Learning: Training Clean Models on Poisoned Data[Code]

    Yige Li, Xixiang Lyu, Nodens Koren, Lingjuan Lyu, Bo Li, Xingjun Ma. NeurIPS, 2021.

  4. Gradient Driven Rewards to Guarantee Fairness in Collaborative Machine Learning

    Xinyi Xu, Lingjuan Lyu, Xingjun Ma, Chenglin Miao, Chuan-Sheng Foo, Kian H. Low. NeurIPS, 2021.

  5. Unlearnable Examples: Making Personal Data Unexploitable [Code] [Webpage]

    Hanxun Huang, Xingjun Ma, Sarah M. Erfani, James Bailey, Yisen Wang. ICLR, 2021. (Spotlight, top 4%) Press: MIT Technology Review, PURSUIT, Gadgets 360.

  6. Neural Attention Distillation: Erasing Backdoor Triggers from Deep Neural Networks [Code]

    Yige Li, Xixiang Lyu, Nodens Koren, Lingjuan Lyu, Bo Li, Xingjun Ma. ICLR, 2021.

  7. Improving Adversarial Robustness via Channel-wise Activation Suppressing [Code]

    Yang Bai, Yuyuan Zeng, Yong Jiang, Shu-Tao Xia, Xingjun Ma, Yisen Wang. ICLR, 2021. (Spotlight, top 4%)

  8. Revisiting Adversarial Robustness Distillation: Robust Soft Labels Make Student Better [Code]

    Bojia Zi*, Shihao Zhao*, Xingjun Ma, Yu-Gang Jiang. ICCV, 2021.

  9. Noise Doesn’t Lie: Towards Universal Detection of Deep Inpainting

    Ang Li, Qiuhong Ke, Xingjun Ma, Haiqin Weng, Zhiyuan Zong, Feng Xue, Rui Zhang. IJCAI, 2021.

  10. Relationships between Local Intrinsic Dimensionality and Tail Entropy [Video]

    James Bailey, Michael Houle, Xingjun Ma. SISAP, Dortmund, Germany, 2021. (Best Paper Award)

  11. RobOT: Robustness-Oriented Testing for Deep Learning Systems [Tookit]

    Jingyi Wang, Jialuo Chen, Youcheng Sun, Xingjun Ma, Dongxia Wang, Jun Sun, Peng Cheng. ICSE, 2021.

  12. Sub-trajectory Similarity Join with Obfuscation

    Yanchuan Chang, Jianzhong Qi, Egemen Tanin, Xingjun Ma, Hanan Samet. SSDBM, 2021. (Best Paper Runner-up Award)

  13. SpineOne: A One-Stage Detection Framework for Degenerative Discs and Vertebrae

    Jiabo He, Wei Liu, Yu Wang, Xingjun Ma, Xian-Sheng Hua. BIBM, 2021.

  14. Dual Head Adversarial Training [Code]

    Yujing Jiang, Xingjun Ma, Sarah M. Erfani, James Bailey. IJCNN, 2021.

  15. Neural Architecture Search via Combinatorial Multi-Armed Bandit

    Hanxun Huang, Xingjun Ma, Sarah M. Erfani, James Bailey. IJCNN, 2021.

  16. Federated Learning with Extreme Label Skew: A Data Extension Approach

    Saheed Tijani, Xingjun Ma, Frank Jiang, Robin Doss. IJCNN, 2021.

  17. Microwave Link Failures Prediction via LSTM-based Feature Fusion Network

    Zichan Ruan, Shuiqiao Yang, Lei Pan, Xingjun Ma, Wei Luo, Marthie Grobler. IJCNN, 2021.

  18. Anomaly Detection for Scenario-based Insider Activities using CGAN Augmented Data

    R G Gayathri, Atul Sajjanhar, Yong Xiang, Xingjun Ma. TrustCom, 2021.

  19. ECG-Adv-GAN: Detecting ECG Adversarial Examples with Conditional Generative Adversarial Networks

    Khondker Fariha Hossain, Sharif Amit Kamran, Alireza Tavakkoli, Lei Pan, Xingjun Ma, Sutharshan Rajasegarar, Chandan Karmaker. ICMLA, 2021.

  20. Exploring the Vulnerability of Natural Language Processing Models via Universal Adversarial Texts

    Xinzhe Li, Ming Liu, Xingjun Ma, Longxiang Gao. ALTA, 2021.

  21. Surgical approach to the facial recess influences the acceptable trajectory of cochlear implantation electrodes

    Bridget Copson, Sudanthi Wijewickrema, Xingjun Ma, Yun Zhou, Jean-Marc Gerard, Stephen O’Leary. European Archives of Oto-Rhino-Laryngology, 1-11, 2021.

2020

  1. Normalized Loss Functions for Deep Learning with Noisy Labels [Code]

    Xingjun Ma*, Hanxun Huang*, Yisen Wang, Simone Romano, Sarah M. Erfani, James Bailey. ICML, 2020.

  2. Improving Adversarial Robustness Requires Revisiting Misclassified Examples [Code]

    Yisen Wang*, Difan Zou*, Jinfeng Yi, James Bailey, Xingjun Ma, Quanquan Gu. ICLR, 2020.

  3. Skip Connections Matter: on the Transferability of Adversarial Examples Generated with ResNets [Code]

    Dongxian Wu, Yisen Wang, Shu-Tao Xia, James Bailey, Xingjun Ma. ICLR, 2020. (Spotlight, top 4%)

  4. Understanding Adversarial Attacks on Deep Learning Based Medical Image Analysis Systems[Code]

    Xingjun Ma*, Yuhao Niu*, Lin Gu, Yisen Wang, Yitian Zhao, James Bailey, Feng Lu. PR, 110, 2021, 107332. (accepted in 2020) Press: Computer Vision News

  5. Clean-Label Backdoor Attacks on Video Recognition Models [Code]

    Shihao Zhao, Xingjun Ma, Xiang Zheng, James Bailey, Jingjing Chen, Yu-Gang Jiang. CVPR, 2020.

  6. Adversarial Camouflage: Hiding Physical-World Attacks with Natural Styles [Code]

    Ranjie Duan, Xingjun Ma, Yisen Wang, James Bailey, Kai Qin, Yun Yang. CVPR, 2020.

  7. WildDeepfake: A Challenging Real-World Dataset for Deepfake Detection [Dataset/Code]

    Bojia Zi, Jingjing Chen, Minghao Chang, Xingjun Ma, Yu-Gang Jiang. MM, 2020.

  8. Reflection Backdoor: A Natural Backdoor Attack on Deep Neural Networks [Code]

    Yunfei Liu, Xingjun Ma, James Bailey, Feng Lu. ECCV, 2020.

  9. Short-Term and Long-Term Context Aggregation Network for Video Inpainting

    Ang Li, Shanshan Zhao, Xingjun Ma, Mingming Gong, Jianzhong Qi, Rui Zhang, Dacheng Tao, Ramamohanarao Kotagiri. ECCV, 2020. (Spotlight, top 5%)

  10. Transfer of Automated Performance Feedback Models to Different Specimens in Virtual Reality Temporal Bone Surgery

    Jesslyn Lamtara, Nathan Hanegbi, Benjamin Talks, Sudanthi Wijewickrema, Xingjun Ma, Patorn Piromchai, James Bailey, Stephen O'Leary. AIED, 2020.

  11. Towards Fair and Privacy-Preserving Federated Deep Models [Code] [Medium] [Youtube]

    Lingjuan Lyu, Jiangshan Yu, Karthik Nandakumar, Yitong Li, Xingjun Ma, Jiong Jin, Han Yu, Kee Siong Ng. TPDS. (accepted in 2020)

  12. How to Democratise and Protect AI: Fair and Differentially Private Decentralised Deep Learning

    Lingjuan Lyu, Yitong Li, Karthik, Nandakumar, Jiangshan Yu, Xingjun Ma. TDSC. (accepted in 2020)

2019

  1. On the Convergence and Robustness of Adversarial Training [Code]

    Yisen Wang*, Xingjun Ma*, James Bailey, Jinfeng Yi, Bowen Zhou, Quanquan Gu. ICML, Long Beach, USA, 2019. (Long talk, top 3%)

  2. Symmetric Cross Entropy for Robust Learning with Noisy Labels [Code]

    Yisen Wang*, Xingjun Ma*, Zaiyi Chen, Yuan Luo, Jinfeng Yi, James Bailey. ICCV, Seoul, Korea, 2019.

  3. Black-box Adversarial Attacks on Video Recognition Models [Code]

    Linxi Jiang*, Xingjun Ma*, Shaoxiang Chen, James Bailey, Yu-Gang Jiang. MM, Nice, France, 2019.

  4. Generative Image Inpainting with Submanifold Alignment

    Ang Li, Jianzhong Qi, Rui Zhang, Xingjun Ma, Ramamohanarao Kotagiri. IJCAI, Macao, China, 2019.

  5. Exploiting Patterns to Explain Individual Predictions

    Yunzhe Jia, James Bailey, Ramamohanarao Kotagiri, Christopher Leckie, Xingjun Ma. KAIS. (accepted in 2019)

2018

  1. Characterizing Adversarial Subspaces Using Local Intrinsic Dimensionality [Code]

    Xingjun Ma, Bo Li, Yisen Wang, Sarah M. Erfani, Sudanthi Wijewickrema, Grant Schoenebeck, Dawn Song, Michael E. Houle, James Bailey. ICLR, Vancouver, BC, Canada, 2018, (Oral, top 2%)

  2. Dimensionality-Driven Learning with Noisy Labels [Code]

    Xingjun Ma*, Yisen Wang*, Michael E. Houle, Shuo Zhou, Sarah M. Erfani, Shu-Tao Xia, Sudanthi Wijewickrema, James Bailey. ICML, Stockholm, Sweden, 2018. (Long talk, top 4%)

  3. Iterative Learning with Open-set Noisy Labels

    Yisen Wang, Weiyang Liu, Xingjun Ma, James Bailey, Hongyuan Zha, Le Song, Shu-Tao Xia. CVPR, Salt Lake City, Utah, USA, 2018. (Spotlight, top 6%)

  4. Providing Automated Real-Time Technical Feedback for Virtual Reality Based Surgical Training: Is the Simpler the Better?

    Sudanthi Wijewickrema, Xingjun Ma, Patorn Piromchai, Robert Briggs, James Bailey, Gregor Kennedy, Stephen O'Leary. AIED, London, UK, 2018.

  5. Development and Validation of a Virtual Reality Tutor to Teach Clinically Oriented Surgical Anatomy of the Ear

    Sudanthi Wijewickrema, Bridget Copson, Xingjun Ma, Robert Briggs, James Bailey, Gregor Kennedy, Stephen O'Leary. CBMS, 2018.

2017

  1. Providing Effective Real-time Feedback in Simulation-based Surgical Training

    Xingjun Ma, Sudanthi Wijewickrema, Yun Zhou, Shuo Zhou, Stephen O'Leary, James Bailey. MICCAI, Quebec City, Canada, 2017.

  2. Adversarial Generation of Real-time Feedback with Neural Networks for Simulation-based Training

    Xingjun Ma, Sudanthi Wijewickrema, Shuo Zhou, Yun Zhou, Zakaria Mhammedi, Stephen O'Leary, James Bailey. IJCAI, Melbourne, Australia, 2017. (Oral)