Yong Li
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Yong Li has a total of 40 co-authors. Here are the top co-authors by number of shared publications:
- Li Duan
Huawei Technologies Düsseldorf
13 shared publications - Lijun Liao
Huawei Technologies Düsseldorf
4 shared publications - Yulian Sun
Huawei Technologies Düsseldorf
4 shared publications - Dong Han
Data Protection Technology Lab Huawei Technologies Düsseldorf Germany
2 shared publications - Joachim Denzler
Friedrich Schiller University Jena
2 shared publications - Asja Fischer
Ruhr University Bochum
1 shared publication
Latest Publications
- 8 October 2026
- Published by Springer Nature
- p. 44-62
0-RTT Integrated in SPDM for Large-Scale AI Super-Clusters
- 8 October 2026
- Published by Springer Nature
- p. 44-62
0-RTT Integrated in SPDM for Large-Scale AI Super-Clusters
- Open Access
- 4 June 2026
- Published by Association for Computing Machinery (ACM)
- p. 661-675
Completing Policy-based Anonymous Tokens: Private Bits, Public Metadata and more…
Anonymous tokens are cryptographic protocols for restricting the access to online resources to eligible users. After proving eligibility to the token issuer, the client receives a set of tokens. Later, it proves eligibility to a resource provider by sending one of its tokens. The protocol ensures that the resource provider cannot link received tokens to their issuance. Recently, Faut et al. (EuroS&P'25) introduced policy-based anonymous tokens, in which a client obtains a single pre-token and locally derives additional tokens according to a publicly announced policy. The major advantage of policy-based tokens is that the communication complexity of the issuance phase is constant. While the work of Faut et al. constitutes a promising step into a new direction, their protocol lacks several desirable properties known from standard anonymous tokens - most notably, the ability to bind a pre-token and all tokens derived from it to a private metadata bit or a publicly known metadata string. In this work, we present a new framework for policy-based anonymous token schemes. Our framework includes two concretely practical constructions - one based on equivalence class signatures and one on algebraic MACs - as well as a communication-optimized, though less practical, construction based on zkSNARKs. All three constructions can be configured to support private metadata bits, public metadata, or both. We formalize the notion of policy-based anonymous tokens with a private metadata bit and public metadata, and we prove security of the two primary constructions: the equivalence-class-signature-based scheme and the algebraic-MAC-based scheme. Finally, we provide an experimental evaluation and comparison of all our constructions alongside the most relevant related work. Our results demonstrate that our two primary constructions achieve significant efficiency improvements over the scheme of Faut et al.
- 4 June 2026
- Published by Association for Computing Machinery (ACM)
- p. 661-675
Open AccessCompleting Policy-based Anonymous Tokens: Private Bits, Public Metadata and more…
Anonymous tokens are cryptographic protocols for restricting the access to online resources to eligible users. After proving eligibility to the token issuer, the client receives a set of tokens. Later, it proves eligibility to a resource provider by sending one of its tokens. The protocol ensures that the resource provider cannot link received tokens to their issuance. Recently, Faut et al. (EuroS&P'25) introduced policy-based anonymous tokens, in which a client obtains a single pre-token and locally derives additional tokens according to a publicly announced policy. The major advantage of policy-based tokens is that the communication complexity of the issuance phase is constant. While the work of Faut et al. constitutes a promising step into a new direction, their protocol lacks several desirable properties known from standard anonymous tokens - most notably, the ability to bind a pre-token and all tokens derived from it to a private metadata bit or a publicly known metadata string. In this work, we present a new framework for policy-based anonymous token schemes. Our framework includes two concretely practical constructions - one based on equivalence class signatures and one on algebraic MACs - as well as a communication-optimized, though less practical, construction based on zkSNARKs. All three constructions can be configured to support private metadata bits, public metadata, or both. We formalize the notion of policy-based anonymous tokens with a private metadata bit and public metadata, and we prove security of the two primary constructions: the equivalence-class-signature-based scheme and the algebraic-MAC-based scheme. Finally, we provide an experimental evaluation and comparison of all our constructions alongside the most relevant related work. Our results demonstrate that our two primary constructions achieve significant efficiency improvements over the scheme of Faut et al.
- Open Access
Realistic Face Reconstruction from Facial Embeddings via Diffusion Models
With the advancement of face recognition (FR) systems, privacy-preserving face recognition (PPFR) systems have gained popularity for their accurate recognition, enhanced facial privacy protection, and robustness to various attacks. However, there are limited studies to further verify privacy risks by reconstructing realistic high-resolution face images from embeddings of these systems, especially for PPFR. In this work, we propose the face embedding mapping (FEM), a general framework that explores Kolmogorov-Arnold Network (KAN) for conducting the embedding-to-face attack by leveraging pre-trained Identity-Preserving diffusion model against state-of-the-art (SOTA) FR and PPFR systems. Based on extensive experiments, we verify that reconstructed faces can be used for accessing other real-word FR systems. Besides, the proposed method shows the robustness in reconstructing faces from the partial and protected face embeddings. Moreover, FEM can be utilized as a tool for evaluating safety of FR and PPFR systems in terms of privacy leakage. All images used in this work are from public datasets.
Open AccessRealistic Face Reconstruction from Facial Embeddings via Diffusion Models
With the advancement of face recognition (FR) systems, privacy-preserving face recognition (PPFR) systems have gained popularity for their accurate recognition, enhanced facial privacy protection, and robustness to various attacks. However, there are limited studies to further verify privacy risks by reconstructing realistic high-resolution face images from embeddings of these systems, especially for PPFR. In this work, we propose the face embedding mapping (FEM), a general framework that explores Kolmogorov-Arnold Network (KAN) for conducting the embedding-to-face attack by leveraging pre-trained Identity-Preserving diffusion model against state-of-the-art (SOTA) FR and PPFR systems. Based on extensive experiments, we verify that reconstructed faces can be used for accessing other real-word FR systems. Besides, the proposed method shows the robustness in reconstructing faces from the partial and protected face embeddings. Moreover, FEM can be utilized as a tool for evaluating safety of FR and PPFR systems in terms of privacy leakage. All images used in this work are from public datasets.
- 20 June 2025
- Published by Springer Nature
- p. 391-414
Obfuscation for Deep Neural Networks Against Model Extraction: Attack Taxonomy and Defense Optimization
- 20 June 2025
- Published by Springer Nature
- p. 391-414
Obfuscation for Deep Neural Networks Against Model Extraction: Attack Taxonomy and Defense Optimization
- 3 May 2025
- Published by Springer Nature
- p. 173-192
Privacy-Preserving Certificate-Less Authenticated Key Exchange with Key Registration Privacy
- 3 May 2025
- Published by Springer Nature
- p. 173-192
Privacy-Preserving Certificate-Less Authenticated Key Exchange with Key Registration Privacy
- 1 April 2025
- Published by Springer Nature
- p. 29-38
Short Paper: PSY: Posterior Sampling Based Privacy Enhancer in Large Language Models
- 1 April 2025
- Published by Springer Nature
- p. 29-38
Short Paper: PSY: Posterior Sampling Based Privacy Enhancer in Large Language Models
- 1 February 2025
- Published by Springer Nature
- p. 106-126
Compressed Cookies: Practical Wildcard Symmetric Searchable Encryption with Optimized Storage
- 1 February 2025
- Published by Springer Nature
- p. 106-126
Compressed Cookies: Practical Wildcard Symmetric Searchable Encryption with Optimized Storage
- Open Access
- 2 December 2024
- Published by Association for Computing Machinery (ACM)
- p. 4718-4731
Protoss: Protocol for Tight Optimal Symmetric Security
We present Protoss, a new balanced PAKE protocol with optimal communication efficiency. Messages are only 160 bits long and the computational complexity is lower than all previous approaches. Our protocol is proven secure in the random oracle model and features a security proof in a strong security model with multiple parties and multiple sessions while allowing for generous attack queries including multiple Test-queries. Moreover, the proof is in the practically relevant single-bit model (that is harder to achieve than the multiple-bit model) and tightly reduces to the Strong Square Diffie-Hellman assumption (SSQRDH). This allows for very efficient, theoretically-sound instantiations and tight compositions with symmetric primitives.
- 2 December 2024
- Published by Association for Computing Machinery (ACM)
- p. 4718-4731
Open AccessProtoss: Protocol for Tight Optimal Symmetric Security
We present Protoss, a new balanced PAKE protocol with optimal communication efficiency. Messages are only 160 bits long and the computational complexity is lower than all previous approaches. Our protocol is proven secure in the random oracle model and features a security proof in a strong security model with multiple parties and multiple sessions while allowing for generous attack queries including multiple Test-queries. Moreover, the proof is in the practically relevant single-bit model (that is harder to achieve than the multiple-bit model) and tightly reduces to the Strong Square Diffie-Hellman assumption (SSQRDH). This allows for very efficient, theoretically-sound instantiations and tight compositions with symmetric primitives.
- Open Access
PSY: Posterior Sampling Based Privacy Enhancer in Large Language Models
Privacy vulnerabilities in LLMs, such as leakage from memorization, have been constantly identified, and various mitigation proposals have been proposed. LoRA is usually used in fine-tuning LLMs and a good entry point to insert privacy-enhancing modules. In this ongoing research, we introduce PSY, a Posterior Sampling based PrivacY enhancer that can be used in LoRA. We propose a simple yet effective realization of PSY using posterior sampling, which effectively prevents privacy leakage from intermediate information and, in turn, preserves the privacy of data owners. We evaluate LoRA extended with PSY against state-of-the-art membership inference and data extraction attacks. The experiments are executed on three different LLM architectures fine-tuned on three datasets with LoRA. In contrast to the commonly used differential privacy method, we find that our proposed modification consistently reduces the attack success rate. Meanwhile, our method has almost no negative impact on model fine-tuning or final performance. Most importantly, PSY reveals a promising path toward privacy enhancement with latent space extensions.
Open AccessPSY: Posterior Sampling Based Privacy Enhancer in Large Language Models
Privacy vulnerabilities in LLMs, such as leakage from memorization, have been constantly identified, and various mitigation proposals have been proposed. LoRA is usually used in fine-tuning LLMs and a good entry point to insert privacy-enhancing modules. In this ongoing research, we introduce PSY, a Posterior Sampling based PrivacY enhancer that can be used in LoRA. We propose a simple yet effective realization of PSY using posterior sampling, which effectively prevents privacy leakage from intermediate information and, in turn, preserves the privacy of data owners. We evaluate LoRA extended with PSY against state-of-the-art membership inference and data extraction attacks. The experiments are executed on three different LLM architectures fine-tuned on three datasets with LoRA. In contrast to the commonly used differential privacy method, we find that our proposed modification consistently reduces the attack success rate. Meanwhile, our method has almost no negative impact on model fine-tuning or final performance. Most importantly, PSY reveals a promising path toward privacy enhancement with latent space extensions.
- 6 September 2024
- Published by Springer Nature
- p. 390-409
Exploiting Internal Randomness for Privacy in Vertical Federated Learning
- 6 September 2024
- Published by Springer Nature
- p. 390-409
Exploiting Internal Randomness for Privacy in Vertical Federated Learning