Yao Huang
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Yao Huang has a total of 30 co-authors. Here are the top co-authors by number of shared publications:
- Xingxing Wei
Beihang University
10 shared publications - Yitong Sun
Institute of Artificial Intelligence, Beihang University, Beijing, China
5 shared publications - Yinpeng Dong
Tsinghua University
4 shared publications - Shouwei Ruan
Beihang University
3 shared publications - Maoxun Yuan
Beihang University
2 shared publications - Ruochen Zhang
Beihang University, Beijing, China
2 shared publications
Latest Publications
- Open Access
- 9 November 2026
- Published by Association for Computing Machinery (ACM)
- p. 14703-14711
ThinkingGuard: Decoding Implicit Hazards via Step-by-Step Risk Attribution in Multimodal Large Language Models
- 9 November 2026
- Published by Association for Computing Machinery (ACM)
- p. 14703-14711
Open AccessThinkingGuard: Decoding Implicit Hazards via Step-by-Step Risk Attribution in Multimodal Large Language Models
- 1 November 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Pattern Analysis and Machine Intelligence
- Vol. 48 (11) , 14155-14170
UniNDM: A Unified Noise-Driven Detection and Mitigation Framework Against Sexual Content in Text-to-Image Generation
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.
- 1 November 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Pattern Analysis and Machine Intelligence
- Vol. 48 (11) , 14155-14170
UniNDM: A Unified Noise-Driven Detection and Mitigation Framework Against Sexual Content in Text-to-Image Generation
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.
- 1 May 2026
- Published by Springer Nature in International Journal of Computer Vision
- Vol. 134 (5) , 253
Knowledge-Guided Adversarial Training for Infrared Object Detection via Thermal Radiation Modeling
- 1 May 2026
- Published by Springer Nature in International Journal of Computer Vision
- Vol. 134 (5) , 253
Knowledge-Guided Adversarial Training for Infrared Object Detection via Thermal Radiation Modeling
- Open Access
- 27 October 2025
- Published by Association for Computing Machinery (ACM)
- p. 11462-11471
NDM: A Noise-driven Detection and Mitigation Framework against Implicit Sexual Intentions in Text-to-Image Generation
- 27 October 2025
- Published by Association for Computing Machinery (ACM)
- p. 11462-11471
Open AccessNDM: A Noise-driven Detection and Mitigation Framework against Implicit Sexual Intentions in Text-to-Image Generation
- Open Access
Breaking the Ceiling: Exploring the Potential of Jailbreak Attacks through Expanding Strategy Space
Large Language Models (LLMs), despite advanced general capabilities, still suffer from numerous safety risks, especially jailbreak attacks that bypass safety protocols. Understanding these vulnerabilities through black-box jailbreak attacks, which better reflect real-world scenarios, offers critical insights into model robustness. While existing methods have shown improvements through various prompt engineering techniques, their success remains limited against safety-aligned models, overlooking a more fundamental problem: the effectiveness is inherently bounded by the predefined strategy spaces. However, expanding this space presents significant challenges in both systematically capturing essential attack patterns and efficiently navigating the increased complexity. To better explore the potential of expanding the strategy space, we address these challenges through a novel framework that decomposes jailbreak strategies into essential components based on the Elaboration Likelihood Model (ELM) theory and develops genetic-based optimization with intention evaluation mechanisms. To be striking, our experiments reveal unprecedented jailbreak capabilities by expanding the strategy space: we achieve over 90% success rate on Claude-3.5 where prior methods completely fail, while demonstrating strong cross-model transferability and surpassing specialized safeguard models in evaluation accuracy. The code is open-sourced at: https://github.com/Aries-iai/CL-GSO.
Open AccessBreaking the Ceiling: Exploring the Potential of Jailbreak Attacks through Expanding Strategy Space
Large Language Models (LLMs), despite advanced general capabilities, still suffer from numerous safety risks, especially jailbreak attacks that bypass safety protocols. Understanding these vulnerabilities through black-box jailbreak attacks, which better reflect real-world scenarios, offers critical insights into model robustness. While existing methods have shown improvements through various prompt engineering techniques, their success remains limited against safety-aligned models, overlooking a more fundamental problem: the effectiveness is inherently bounded by the predefined strategy spaces. However, expanding this space presents significant challenges in both systematically capturing essential attack patterns and efficiently navigating the increased complexity. To better explore the potential of expanding the strategy space, we address these challenges through a novel framework that decomposes jailbreak strategies into essential components based on the Elaboration Likelihood Model (ELM) theory and develops genetic-based optimization with intention evaluation mechanisms. To be striking, our experiments reveal unprecedented jailbreak capabilities by expanding the strategy space: we achieve over 90% success rate on Claude-3.5 where prior methods completely fail, while demonstrating strong cross-model transferability and surpassing specialized safeguard models in evaluation accuracy. The code is open-sourced at: https://github.com/Aries-iai/CL-GSO.
- Top Cited PapersOpen Access
- 1 January 2025
- Published by Association for Computational Linguistics (ACL)
- p. 7870-7888
Breaking the Ceiling: Exploring the Potential of Jailbreak Attacks through Expanding Strategy Space
Yao Huang, Yitong Sun, Shouwei Ruan, Yichi Zhang, Yinpeng Dong, Xingxing Wei. Findings of the Association for Computational Linguistics: ACL 2025. 2025.
- 1 January 2025
- Published by Association for Computational Linguistics (ACL)
- p. 7870-7888
Top Cited PapersOpen AccessBreaking the Ceiling: Exploring the Potential of Jailbreak Attacks through Expanding Strategy Space
Yao Huang, Yitong Sun, Shouwei Ruan, Yichi Zhang, Yinpeng Dong, Xingxing Wei. Findings of the Association for Computational Linguistics: ACL 2025. 2025.
- 29 October 2024
- Published by Springer Nature
- p. 309-327
Omniview-Tuning: Boosting Viewpoint Invariance of Vision-Language Pre-training Models
- 29 October 2024
- Published by Springer Nature
- p. 309-327
Omniview-Tuning: Boosting Viewpoint Invariance of Vision-Language Pre-training Models
- Open Access
- 28 October 2024
- Published by Association for Computing Machinery (ACM)
- p. 5902-5910
Embodied Laser Attack:Leveraging Scene Priors to Achieve Agent-based Robust Non-contact Attacks
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.
- 28 October 2024
- Published by Association for Computing Machinery (ACM)
- p. 5902-5910
Open AccessEmbodied Laser Attack:Leveraging Scene Priors to Achieve Agent-based Robust Non-contact Attacks
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.
- 22 December 2023
- Published by Springer Nature in International Journal of Computer Vision
- Vol. 132 (6) , 1928-1944
Infrared Adversarial Patches with Learnable Shapes and Locations in the Physical World
- 22 December 2023
- Published by Springer Nature in International Journal of Computer Vision
- Vol. 132 (6) , 1928-1944
Infrared Adversarial Patches with Learnable Shapes and Locations in the Physical World
- 6 November 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Pattern Analysis and Machine Intelligence
- Vol. 46 (4) , 2348-2363
Unified Adversarial Patch for Visible-Infrared Cross-Modal Attacks in the Physical World
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.
- 6 November 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Pattern Analysis and Machine Intelligence
- Vol. 46 (4) , 2348-2363
Unified Adversarial Patch for Visible-Infrared Cross-Modal Attacks in the Physical World
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.