Profiles2023-2026ORCID 0000-0001-7348-5625
Tsung-Yi Ho
The Chinese University of Hong Kong Sha Tin
13Papers
11Last author
19Unique co-authors
0First author
0Solo
First and last author come from byline position, which some venues assign alphabetically.
Recent papers13
- GRE Score: Generative Risk Evaluation for Large Language ModelsAuthor 3 of 3ACL Findings 2026
- CoP: Agentic Red-teaming for Large Language Models using Composition of PrinciplesAuthor 3 of 3NeurIPS 2025 · 13 citations
- BQSim: GPU-accelerated Batch Quantum Circuit Simulation using Decision DiagramAuthor 4 of 5ASPLOS 2025 · 9 citations
- Defensive Prompt Patch: A Robust and Generalizable Defense of Large Language Models against Jailbreak AttacksAuthor 4 of 4ACL Findings 2025 · 9 citations
- CARE: Decoding-Time Safety Alignment via Rollback and Introspection InterventionAuthor 5 of 5NeurIPS 2025 · 5 citations
- Gradient Cuff: Detecting Jailbreak Attacks on Large Language Models by Exploring Refusal Loss LandscapesAuthor 3 of 3NeurIPS 2024 · 97 citations
- MMA-Diffusion: MultiModal Attack on Diffusion ModelsAuthor 4 of 6CVPR 2024 · 31 citations
- GREAT Score: Global Robustness Evaluation of Adversarial Perturbation using Generative ModelsAuthor 3 of 3NeurIPS 2024 · 8 citations
- RCGP: An Automatic Synthesis Framework for Reversible Quantum-Flux-Parametron Logic Circuits based on Efficient Cartesian Genetic ProgrammingAuthor 3 of 3DAC 2024 · 4 citations
- NeuralFuse: Learning to Recover the Accuracy of Access-Limited Neural Network Inference in Low-Voltage RegimesAuthor 5 of 5NeurIPS 2024 · 2 citations
- RADAR: Robust AI-Text Detection via Adversarial LearningAuthor 3 of 3NeurIPS 2023 · 315 citations
- VillanDiffusion: A Unified Backdoor Attack Framework for Diffusion ModelsAuthor 2 of 2NeurIPS 2023 · 101 citations
- How to Backdoor Diffusion Models?Author 3 of 3CVPR 2023
