HybridPETs 2027 Workshop on Hybrid Privacy-Enhancing Technologies

Co-located with
ACNS 2027 25th International Conference on Applied Cryptography and Network Security
Conference dates
–
Workshop date
TBA
Location
Royal Holloway, University of London Egham, United Kingdom

Call for Papers

Call for Papers

Real-world deployments of privacy-preserving systems increasingly combine several privacy-enhancing technologies (PETs), such as secure multi-party computation (MPC), homomorphic encryption (HE), trusted execution environments (TEEs), differential privacy (DP), zero-knowledge proofs (ZKPs), and private information retrieval (PIR). These techniques protect different aspects of a computation, rely on different trust assumptions, and come with very different costs. Large-scale applications, such as private inference on large language models, often require more than one of them.

HybridPETs is a new full-day workshop at ACNS 2027 dedicated to such combinations. It brings together researchers and practitioners from cryptography, privacy, systems, and hardware security to discuss how heterogeneous PETs can be combined securely and efficiently, which end-to-end guarantees the resulting systems provide, and how they can be implemented, accelerated, verified, and deployed.

Submissions should address the combination of two or more PETs, or of a PET with trusted hardware, hardware acceleration, or other system components. Work on a single PET in isolation, on general AI security, or on hardware acceleration without such a combination is out of scope.

Topics of interest

Topics of interest include, but are not limited to:

  • Protocols and architectures combining multiple PETs, e.g., MPC with HE, HE with ZKPs, PIR with HE, or private set intersection with complementary techniques
  • TEEs combined with cryptographic techniques or DP, including attestation-based protocols and cryptographic fallback after a TEE compromise
  • Secure aggregation, federated learning, and distributed analytics with combined cryptographic and DP guarantees
  • Hybrid PETs for machine learning and AI: private training, fine-tuning, and inference, including large language models, retrieval-augmented generation, and AI agents
  • Selection, orchestration, and switching between PETs, and secure conversion between protected representations
  • Compilers and intermediate representations for heterogeneous PET pipelines
  • Composition of security and privacy guarantees, formal verification, and failure analysis of hybrid PET systems
  • Verifiability and accountability, e.g., verifiable confidential computing, ZKPs combined with privacy mechanisms, and audit evidence
  • Hardware/software co-design and acceleration of hybrid PETs on GPUs, FPGAs, and ASICs, and the interaction between TEEs and accelerators
  • Benchmarking, performance evaluation, and interoperability of hybrid PET systems
  • Applications and deployment experience, e.g., in healthcare, genomics, finance, mobility, digital identity, IoT, threat intelligence, and enterprise AI

Submissions

In addition to new techniques and systems, we welcome rigorous negative results, practical deployment experience, and extensive performance or use-case evaluations.

All submissions must follow the guidelines below.

Format
All submissions must be in PDF format, following the unmodified LNCS format.
Page limits
At most 18 pages excluding references and appendices, and at most 24 pages in total, including references and appendices. Reviewers are not required to read appendices, so submissions should be intelligible without them.
Anonymity
Reviewing is double-blind. Submissions must be anonymous, with no author names, affiliations, acknowledgments, or obvious references. Supplementary artifacts, such as source code or data, may be made available to reviewers through an anonymous repository (e.g., anonymous.4open.science), with the anonymous URL included in the paper.
Originality
Submissions must not substantially overlap with papers that have been published (other than as preprints) or accepted for publication, or that are simultaneously submitted to a journal, conference, or workshop with published proceedings.
Conflicts of interest
As defined in the ACNS 2027 Call for Papers: a shared institutional affiliation at the time of submission (including secondary affiliations and consulting), an advisor–advisee relationship at any time, collaboration or joint publication within the prior two years, sponsorship or administration of a grant funding the research, or close personal or family ties.
Submission site
Submission link to be announced

Submissions that do not follow these guidelines or fall outside the scope of the workshop may be rejected without review at the discretion of the PC chairs.

Publication and presentation

Accepted papers will be published in the joint post-proceedings of the ACNS 2027 workshops in Springer’s LNCS series. Camera-ready versions are expected to be due on 14 June 2027 (AoE). All accepted papers will be considered for the ACNS 2027 Best Workshop Paper Award (EUR 500, sponsored by Springer).

Every accepted workshop paper requires at least one regular (non-student) conference registration and an in-person presentation. ACNS 2027 registration

Important dates

Submission
AoE
Notification
AoE
Camera-ready
AoE
Workshop
TBA
Between 28 June and 1 July 2027

Contact

Organization

PC Chairs

  • Ajith Suresh

    Technology Innovation Institute, UAE

  • Hossein Yalame

    Bosch Research, Germany

Contact:

Web Chair

  • Christian Weinert

    Royal Holloway, University of London, UK

Program Committee

TBA

Location & registration

Location

HybridPETs 2027 is a full-day, in-person workshop held during ACNS 2027 in the Windsor Building at Royal Holloway, University of London, in Egham, United Kingdom. The exact date will be announced on this page.

Information on the venue, travel to the campus, and accommodation is available on the ACNS 2027 website.

Registration

Registration is handled by ACNS 2027. Please refer to the conference’s registration page for categories, fees, and deadlines.

Every accepted workshop paper requires at least one regular (non-student) conference registration and an in-person presentation.