Tinghao Xie

8 papers · 2022–2025 · 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 (11)

Venues

ICLR (6) CVPR (1) ICML (1)

Top co-authors

Xiangyu Qi (7) Prateek Mittal (6) Peter Henderson (5) Yangsibo Huang (4) Boyi Wei (3) Luxi He (3) Ruoxi Jia (2) Yi Zeng (2) Kaixuan Huang (2) Yiming Li (2)

Research topics

Artificial Intelligence (7) Core AI (7) Machine Learning (4) Application Areas (3) Deep Learning (2) Computer Vision (1) Architectures (1) Generation (1) Models (1) Optimization & Theory (1)

Keywords

harmful content (2) safety alignment (2) deep learning (2) adversarial attack (2) cluster analysis (1) video generation (1) model unlearning (1) backdoor attack (1) deep neural network (1) adversarial defense (1) safety evaluation (1) model deployment (1) backdoor detection (1) fine tuning (1) model customization (1) copyright infringement (1) large language model (1) neural network (1) open-weight model (1) prompt rewriting (1)

Papers

Fantastic Copyrighted Beasts and How (Not) to Generate Them ICLR 2025 On Evaluating the Durability of Safeguards for Open-Weight LLMs ICLR 2025 SORRY-Bench: Systematically Evaluating Large Language Model Safety Refusal ICLR 2025 Assessing the Brittleness of Safety Alignment via Pruning and Low-Rank Modifications ICML 2024 BaDExpert: Extracting Backdoor Functionality for Accurate Backdoor Input Detection ICLR 2024 Fine-tuning Aligned Language Models Compromises Safety, Even When Users Do Not Intend To! ICLR 2024 Revisiting the Assumption of Latent Separability for Backdoor Defenses ICLR 2023 Towards Practical Deployment-Stage Backdoor Attack on Deep Neural Networks CVPR 2022