Yitong Sun

Yitong Sun

Auto-generated profile, based on publicly available information

Yitong Sun

Institute of Artificial Intelligence, Beihang University, Beijing, China

Research Focus & Metrics

Academic Career Length

5Years (2022-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
6 November 2023 in IEEE Transactions on Pattern Analysis and Machine Intelligence
Verified
Journal Article
Closed access
XWXingxing Wei
Xingxing Wei
YHYao Huang
Yao Huang
YSYitong Sun
Yitong Sun
JYJie Yu
Jie Yu
Physical adversarial attacks have put a severe threat to DNN-based object detectors. To enhance security, a combination of visible and infrared sensors is deployed in various scenarios, which has proven effective in disabling existing single-modal physical attacks. To further demonstrate the potential risks in such cases, we design a unified adversarial patch that can perform cross-modal physical attacks, achieving evasion in both modalities simultaneously with a single patch. Given the different imaging mechanisms of visible and infrared sensors, our work manipulates patches' shape features, which can be captured in different modalities when they undergo changes. To deal with challenges, we propose a novel boundary-limited shape optimization approach that aims to achieve compact and smooth shapes for the adversarial patch, making it easy to implement in the physical world. And a score-aware iterative evaluation method is also introduced to balance the fooling degree between visible and infrared detectors during optimization, which guides the adversarial patch to iteratively reduce the predicted scores of the multi-modal sensors. Furthermore, we propose an Affine-Transformation-based enhancement strategy that makes the learnable shape robust to various angles, thus mitigating the issue of shape deformation caused by different shooting angles in the real world. Our method is evaluated against several state-of-the-art object detectors, achieving an Attack Success Rate (ASR) of over 80%. We also demonstrate the effectiveness of our approach in physical-world scenarios under various settings, including different angles, distances, postures, and scenes for both visible and infrared sensors.

28 October 2024
Verified
Conference Paper
Open access
YSYitong Sun
Yitong Sun
YHYao Huang
Yao Huang
XWXingxing Wei
Xingxing Wei
As physical adversarial attacks become extensively applied in unearthing the potential risk of security-critical scenarios, especially in dynamic scenarios, their vulnerability to environmental variations has also been brought to light. The non-robust nature of physical adversarial attack methods brings less-than-stable performance consequently. Although methods such as Expectation over Transformation (EOT) have enhanced the robustness of traditional contact attacks like adversarial patches, they fall short in practicality and concealment within dynamic environments such as traffic scenarios. Meanwhile, non-contact laser attacks, while offering enhanced adaptability, face constraints due to a limited optimization space for their attributes, rendering EOT less effective. This limitation underscores the necessity for developing a new strategy to augment the robustness of such practices. To address these issues, this paper introduces the Embodied Laser Attack (ELA), a novel framework that leverages the embodied intelligence paradigm of Perception-Decision-Control to dynamically tailor non-contact laser attacks. For the perception module, given the challenge of simulating the victim's view by full-image transformation, ELA has innovatively developed a local perspective transformation network, based on the intrinsic prior knowledge of traffic scenes and enables effective and efficient estimation. For the decision and control module, ELA trains an attack agent with data-driven reinforcement learning instead of adopting time-consuming heuristic algorithms, making it capable of instantaneously determining a valid attack strategy with the perceived information by well-designed rewards, which is then conducted by a controllable laser emitter. Experimentally, we apply our framework to diverse traffic scenarios both in the digital and physical world, verifying the effectiveness of our method under dynamic successive scenes.


1 June 2023
Verified
Conference Paper
Closed access
XDXiaoqiang Di
Xiaoqiang Di
YSYitong Sun
Yitong Sun
JLJunchu Lu
Junchu Lu
WDWeiyi Dai
Weiyi Dai
HQHui Qi
Hui Qi
Internet of Vehicles (IoV) has developed into a useful service platform between vehicles and other vehicles, vehicles and roadways, and vehicles and pedestrians as a result of advancements in communication technology. However, along with the intelligent and networked IoV, IoV security issues are increasingly emerging, which often lead to extremely serious consequences. For example, when a vehicle is remotely attacked and maliciously controlled, it can lead to user privacy leakage. Attackers can also control the driving of vehicles through attacks, and there may even be a large number of connected cars being controlled in bulk, ultimately causing major social security incidents. In the Internet of Vehicles, authentication technology is one of the main solutions to address the security issues of the Internet of Vehicles. Many researchers have conducted some research on the authentication technology of vehicular network, but the traditional authentication scheme under the centralized system has problems such as frequent single-point failures and high overhead. When blockchain technology is properly implemented to the vehicle network environment, its security outperforms that of centralized systems due to features like decentralization, tamper-proof, security, and traceability. Because of this, we construct an optimization technique based on the Practical Byzantine Fault Tolerance (PBFT) algorithm in this work and present an efficient IoV decentralized authentication scheme based on the consensus algorithm of blockchain technology. After performance analysis, it is proved that our scheme has better performance and higher security.

8 June 2022 in arXiv
Verified
Preprint
Open access
PDF file
MZMingtian Zhang
Mingtian Zhang
AZAndi Zhang
Andi Zhang
TXTim Z. Xiao
Tim Z. Xiao
YSYitong Sun
Yitong Sun
SMSteven McDonagh
Steven McDonagh
Density-based Out-of-distribution (OOD) detection has recently been shown unreliable for the task of detecting OOD images. Various density ratio based approaches achieve good empirical performance, however methods typically lack a principled probabilistic modelling explanation. In this work, we propose to unify density ratio based methods under a novel framework that builds energy-based models and employs differing base distributions. Under our framework, the density ratio can be viewed as the unnormalized density of an implicit semantic distribution. Further, we propose to directly estimate the density ratio of a data sample through class ratio estimation. We report competitive results on OOD image problems in comparison with recent work that alternatively requires training of deep generative models for the task. Our approach enables a simple and yet effective path towards solving the OOD detection problem.

1 November 2026 in IEEE Transactions on Pattern Analysis and Machine Intelligence
Verified
Journal Article
Closed access
YHYao Huang
Yao Huang
YSYitong Sun
Yitong Sun
HCHuanran Chen
Huanran Chen
RZRuochen Zhang
Ruochen Zhang
SRShouwei Ruan
Shouwei Ruan
XWXingxing Wei
Xingxing Wei
Despite the impressive generative capabilities of text-to-image (T2I) diffusion models, they remain vulnerable to implicit sexual prompts, where subtle cues disguised as benign terms or adversarial tokens unexpectedly generate the inappropriate content due to model biases or latent correlations in training data. Existing safety mechanisms face fundamental limitations: detection methods primarily identify explicit content and fail to capture implicit malicious intent, while mitigation approaches rely on static negative prompts inadequate for diverse implicit scenarios. To address these challenges, we propose UniNDM, a unified noise-driven framework that rethinks safety mechanisms through the lens of noise dynamics in diffusion processes. Our key insight is that early-stage predicted noise exhibits inherent separability between normal and sexually explicit content, which we theoretically demonstrates quadratically increasing semantic concentration with timestep. Leveraging this property, we develop a lightweight noise-based detector achieving superior accuracy with virtually no computational overhead. For mitigation, we introduce noise-enhanced adaptive negative guidance: dynamically generating context-specific negative prompts via large language models to handle diverse implicit content, while optimizing initial noise by suppressing attention concentration on explicit tokens to provide comprehensive protection. Besides the U-Net-based diffusion models, we further extend our framework to emerging Diffusion Transformer architectures through region-constrained semantic guidance tailored for their unified multimodal attention. Comprehensive experiments across U-Net models (SDv1.4, v1.5, v2.1, XL) and DiT models (SDv3) on both natural and adversarial datasets demonstrate substantial improvements over state-of-the-art methods, including SLD, UCE, Safree, etc. Our code is publicly available at https://github.com/Aries-iai/UniNDM.