Profiles2020-2026
Yinpeng Dong
Tsinghua University
31Papers
1Last author
44Unique co-authors
4First author
0Solo
First and last author come from byline position, which some venues assign alphabetically.
Recent papers31
- Unveiling the Basin-Like Loss Landscape in Large Language ModelsAuthor 5 of 6ICLR 2026 · 14 citations
- Silent Leaks: Implicit Knowledge Extraction Attack on RAG SystemsAuthor 7 of 8ICLR 2026 · 13 citations
- Towards Safe Reasoning in Large Reasoning Models via Corrective InterventionAuthor 9 of 10ICLR 2026 · 13 citations
- Why Do Unlearnable Examples Work: A Novel Perspective of Mutual InformationAuthor 3 of 4ICLR 2026 · 3 citations
- Mitigating Overthinking in Large Reasoning Models via Manifold SteeringAuthor 6 of 6NeurIPS 2025 · 45 citations
- Breaking the Ceiling: Exploring the Potential of Jailbreak Attacks through Expanding Strategy SpaceAuthor 5 of 6ACL Findings 2025 · 8 citations
- Efficient Input-Level Backdoor Defense on Text-to-Image Synthesis via Neuron Activation VariationAuthor 9 of 10ICCV 2025 · 2 citations
- AdvDreamer Unveils: Are Vision-Language Models Truly Ready for Real-World 3D Variations?Author 7 of 8ICCV 2025
- STAIR: Improving Safety Alignment with Introspective ReasoningAuthor 9 of 10ICML 2025
- Rethinking Model Ensemble in Transfer-based Adversarial AttacksAuthor 3 of 6ICLR 2024 · 112 citations
- Robust Classification via a Single Diffusion ModelAuthor 2 of 7ICML 2024 · 94 citations
- Membership Inference on Text-to-Image Diffusion Models via Conditional Likelihood DiscrepancyAuthor 3 of 8NeurIPS 2024 · 48 citations
- Efficient Black-box Adversarial Attacks via Bayesian Optimization Guided by a Function PriorAuthor 3 of 6ICML 2024 · 15 citations
- Improving Robustness of 3D Point Cloud Recognition from a Fourier PerspectiveAuthor 2 of 6NeurIPS 2024 · 15 citations
- Toward Availability Attacks in 3D Point CloudsAuthor 3 of 4ICML 2024 · 9 citations
- Embodied Active Defense: Leveraging Recurrent Feedback to Counter Adversarial PatchesAuthor 3 of 6ICLR 2024 · 6 citations
- Towards Transferable Targeted 3D Adversarial Attack in the Physical WorldAuthor 2 of 6CVPR 2024
- Text-to-Image Diffusion Models can be Easily Backdoored through Multimodal Data PoisoningAuthor 2 of 6ACM MM 2023 · 46 citations
- Towards Viewpoint-Invariant Visual Recognition via Adversarial TrainingAuthor 2 of 6ICCV 2023 · 23 citations
- Learning Sample Difficulty from Pre-trained Models for Reliable PredictionAuthor 4 of 5NeurIPS 2023 · 21 citations
- Root Pose Decomposition Towards Generic Non-rigid 3D Reconstruction with Monocular VideosAuthor 2 of 4ICCV 2023 · 11 citations
- Benchmarking Robustness of 3D Object Detection to Common Corruptions in Autonomous DrivingAuthor 1 of 9CVPR 2023
- Compacting Binary Neural Networks by Sparse Kernel SelectionAuthor 3 of 5CVPR 2023
- Towards Effective Adversarial Textured 3D Meshes on Physical Face RecognitionAuthor 5 of 8CVPR 2023
- Understanding the Robustness of 3D Object Detection with Bird'View Representations in Autonomous DrivingAuthor 4 of 8CVPR 2023
- ViewFool: Evaluating the Robustness of Visual Recognition to Adversarial ViewpointsAuthor 1 of 6NeurIPS 2022 · 72 citations
- Isometric 3D Adversarial Examples in the Physical WorldAuthor 2 of 4NeurIPS 2022 · 45 citations
- Black-box Detection of Backdoor Attacks with Limited Information and DataAuthor 1 of 7ICCV 2021 · 128 citations
- Towards Face Encryption by Generating Adversarial Identity MasksAuthor 2 of 7ICCV 2021 · 109 citations
- Boosting Adversarial Training with Hypersphere EmbeddingAuthor 3 of 6NeurIPS 2020 · 170 citations
Showing the 30 most recent indexed papers.
