Tianwei Zhang
Associate Professor
Works on computer system and AI security, including trustworthy machine learning and the safety of large models.
Tianwei Zhang's lab at NTU works on trustworthy & safe ML, interpretability, evaluation & safety of LMs and data management & big data. 211 papers since 2023, 17 at top venues, mostly USENIX Security and TPAMI. 12 PhD students. 14 alumni.
At a glance
- papers, 2023–2026
- 55/ year
- at T1 venues
- 8%
- PhD students
- 12
- MS students
- 0
- alumni listed
- 14
Papers by yearfrom the lab homepage, 2023–2026
- 58 papers2026 · T1 6 · collecting
- 63 papers2025 · T1 5
- 58 papers2024 · T1 3
- 32 papers2023 · T1 3
T1 venuesOtherVenues: USENIX Security, TPAMI, TCSVT, TIP, TMM
Research focuspaper titles that use each topic's words
- Interpretability, evaluation & safety of LMs8 papers
- Data management & big data8 papers
- system6 papers
- Trustworthy & safe ML3 papers
- trustworthy machine0 papers
Paper list
From the lab homepage.
202658 papers
- IEEE International Symposium on Hardware Oriented Security and Trust (HOST), MayAdvProtego: A Unified Framework to SafeGuard Deep Learning Models Against Side Channel Leakage
- IEEE Transactions on Dependable and Secure Computing, Volume: 23, Issue: 3, MayAn Advanced Gradient Leakage Attack against Duplicate Labels via Model Output Reconstruction
- International Conference on Learning Representations (ICLR), AprilAudioTrust: Benchmarking The Multifaceted Trustworthiness of Audio Large Language Models
- Conference on Empirical Methods in Natural Language Processing (EMNLP), OctoberAuditing Chinese Web-scale Corpora via Sampled BPE Token Statistics
- International Conference on Learning Representations (ICLR), AprilAutoregressive-based Progressive Coding for Ultra-Low Bitrate Image Compression
- ACM International Conference on Multimedia (MM), OctoberBeyond Retrieval: Improving Evidence Quality for LLM-based Multimodal Fact-Checking
- IEEE/ACM International Conference on Computer-Aided Design (ICCAD), NovemberBitHydra: Towards Bit-flip Inference Cost Attack against Large Language Model
- International Conference on Machine Learning (ICML), JulyCONCUR: High-Throughput Agentic Batch Inference of LLM via Congestion-Based Concurrency Control
- Neural Information Processing Systems (NeurIPS), DecemberDAWN: Dependency-Aware Fast Inference for Diffusion LLMs
- Neural Information Processing Systems (NeurIPS), DecemberDECEIVE-AFC: Adversarial Claim Attacks against Search-Enabled LLM-based Fact-Checking Systems
- USENIX Symposium on Networked Systems Design and Implementation (NSDI), MayDi-PS: System-Algorithm Co-Design for Asynchronous and Heterogeneous Cross-cluster LLM Training at Scale
- ACM Transactions on Software Engineering and MethodologyDiscovering Safety Violations of Decision-Making in Autonomous Driving Systems from Accident-Free Traffic Scenarios
202563 papers
- International Conference on Learning Representations (ICLR), AprilA Benchmark for Semantic Sensitive Information in LLMs Outputs
- IEEE Transactions on Intelligent Vehicles, Volume: 10, Issue: 9, SeptemberAdversarial Attacks on Autonomous Driving Systems in the Physical World: a Survey
- IEEE Transactions on Forensics and Security, Volume: 21, DecemberAlkaid: Accelerating Three-Party Boolean Circuits by Mixing Correlations and Redundancy
- International Conference on Learning Representations (ICLR), AprilAn Engorgio Prompt Makes Large Language Model Babble on
- AAAI Conference on Artificial Intelligence (AAAI), Oral, FebruaryAn LLM-empowered Adaptive Evolutionary Algorithm For Multi-Component Deep Learning Systems
- IEEE Transactions on Dependable and Secure Computing, Volume: 22, Issue: 6, NovemberAntelope: Fast and Secure Neural Network Inference
- IEEE International Conference on Multimedia & Expo (ICME), JuneAparecium: Revealing Secrets from Physical Photographs
- IEEE Transactions on Vehicular Technology, Volume: 74, Issue: 10, OctoberAuthUp: A Privacy-Preserving V2I Authentication Scheme for Vehicular Ad Hoc Networks
- IEEE International Conference on Computer Vision (ICCV), OctoberAutomated Red Teaming for Text-to-Image Models through Feedback-Guided Prompt Iteration with Vision-Language Models
- International Conference on Machine Learning (ICML), JulyBSemiFL: Semi-supervised Federated Learning via a Bayesian Approach
- IEEE Symposium on Security and Privacy (S&P), MayCamLoPA: A Hidden Wireless Camera Localization Framework via Signal Propagation Path Analysis
- TMMClean Image May be Dangerous: Data Poisoning Attacks Against Deep Hashing
202458 papers
- IEEE Transactions on Dependable and Secure Computing, Volume: 21, Issue: 2, MarchA Comprehensive Defense Framework against Model Extraction Attacks
- IEEE Transactions on Intelligent Transportation Systems, Volume: 25, Issue: 3, MarchA Tamper-resistant Broadcasting Scheme for Secure Communication in Internet of Autonomous Vehicles
- IEEE Transactions on Computers, Volume: 73, Issue: 3, MarchAn Efficient Preprocessing-based Approach to Mitigate Advanced Adversarial Attacks
- International Conference on Machine Learning (ICML), JulyAquaLoRA: Toward White-box Protection for Customized Stable Diffusion Models via Watermark LoRA
- Neural Information Processing Systems (NeurIPS), DecemberART: Automatic Red-teaming for Text-to-Image Models to Protect Benign Users
- ACM International Conference on Supercomputing (ICS), JuneAutoSched: An Adaptive Self-configured Framework for Scheduling Deep Learning Training Workloads
- European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases (ECML-PKDD), SeptemberBackdoor Attacks with Input-unique Triggers in NLP
- IEEE Symposium on Security and Privacy (S&P), MayBackdooring Multimodal Learning
- International Conference on Learning Representations (ICLR), MayBadEdit: Backdooring Large Language Models by Model Editing
- Neural Information Processing Systems (NeurIPS), DecemberBeware of Road Markings: A New Adversarial Patch Attack to Monocular Depth Estimation
- IEEE Transactions on Information Forensics and Security, Volume: 19, AprilBoosting Black-box Attack to Deep Neural Networks with Conditional Diffusion Models
- ACM ASIA Conference on Computer and Communications Security (AsiaCCS), JulyCapsuleFormer: A Capsule and Transformer Combined Model for Decentralized Application Encrypted Traffic Classification
202332 papers
- IEEE Transactions on Intelligent Transportation Systems, Volume: 24, Issue: 1, JanuaryADS-Lead: Lifelong Anomaly Detection in Autonomous Driving Systems
- USENIX Security (T1)Aegis: Mitigating Targeted Bit-flip Attacks against Deep Neural Networks
- IEEE International Conference on Data Mining (ICDM), DecemberATTA: Adversarial Task-transferable Attacks on Autonomous Driving Systems
- TPAMI (T1)Automatic Transformation Search Against Deep Leakage from Gradients
- Volume: 20, Issue: 4, JulyCan We Mitigate Backdoor Attack Using Adversarial Detection Methods
- International Conference on Learning Representations (ICLR), Notable-top-5%, MayClean-image Backdoor: Attacking Multi-label Models with Poisoned Labels Only
- IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), JuneColor Backdoor: A Robust Poisoning Attack in Color Space
- IEEE International Conference on Computer Vision (ICCV), OctoberComputation and Data Efficient Backdoor Attacks
- IEEE International Conference on Computer Vision (ICCV), Oral, OctoberDeep Multitask Learning with Progressive Parameter Sharing
- AAAI Conference on Artificial Intelligence (AAAI), FebruaryDefending Against Backdoor Attacks in Natural Language Generation
- IEEE Transactions on Information Forensics and Security, Volume: 19, SeptemberDifferentially Private Federated Learning with an Adaptive Noise Mechanism
- IEEE Transactions on Information Forensics and Security, Volume: 18, AugustErase and Repair: An Efficient Box-Free Removal Attack on High-Capacity Deep Hiding
Where alumni went14 listed
Apply to NTUFall 2027
Not announced yet
Computing and Data Science PhD
- Round
- Main
- Opens
- —
- GRE
- not required
- English
- TOEFL/IELTS
- Fee
- See program page
Not yet confirmed for this intake.
Similar labsnearest by research summary
- USUC Berkeley
Nika Haghtalab
- Learning theory & optimization
- CNSJTU
Hongsheng Hu
- Trustworthy & safe ML
- USUCSD
Parinaz Naghizadeh
- Reasoning & neuro-symbolic AI
- USUCSD
Tsui-Wei Weng
- Trustworthy & safe ML
- USCMU
Vincent Conitzer
- Reasoning & neuro-symbolic AI
- USUT Austin
Kaushik Chowdhury
- Trustworthy & safe ML
Your notes
Save this lab to keep notes.
Data
Checked against the lab homepage.
- Collected
- Verified
- Papers
- The lab homepage
- Members
- Lab homepage
- Alumni
- Lab homepage
Tiers are venue classifications, not lab rankings.