Publications
Our publications from 2019 onwards can be found on the CrySP publications page. Earlier publications are available on the Aalto SSG publications page.
Sometimes we blog about our research results.
Events
- “Systems Security Research at KTH” workshop (24 April 2026)
Software
Below is a list of publicly available source code related to past and current SSG projects.
- Extracting Training Data from Diffusion Language Models via Infilling (2026)
- Locket: Robust feature-locking technique for language models (2026)
- SoK: Colluding adversaries in machine learning pipelines (2026)
- PICASSO: Scaling CHERI use-after-free protection using colored capabilities (2026)
- BLACKOUT (2025)
- Combining machine learning defenses without conflicts (2025)
- Espresso (2025)
- Laminator: Verifiable ML property cards using hardware-assisted attestations (2025)
- Amulet (2024)
- Distributional Attestation (2024)
- Code for SoK: Unintended Interactions among ML Defenses (2024)
- False Claims against Model Ownership Resolution (2024)
- GLUE (2024)
- BliMe: Blinded Memory (2024)
- BliMe formal model in F* (2024)
- SeMalloc: Semantics-informed memory allocator (2024)
- S2malloc: Statistically secure allocator for use-after-free protection (2024)
- GrOVe: Ownership Verification of Graph Neural Networks using Embeddings (2024)
- CONF-ML: Conflicts between ML protection mechanisms (2023)
- FIA Fingerprinting (2023)
- FLARE: Fingerprinting Deep Reinforcement Learning Agents using Universal Adversarial Masks (2023)
- AD3: Adversarial input detector for deep reinforcement learning (DRL) (2022)
- Pointer Authentication: Projects related to ARMv8.3-PAuth (2021)
- DAWN: Dynamic adversarial watermarking of neural networks (2021)
- ParChoice: Effective writing style transfer via combinatorial paraphrasing (2020)
- WAFFLE: Watermarking in federated learning (2020)
- Language data augmentation: Data augmentation techniques for toxic language classification (2020)
- RecAgglo: Recursive Agglomerative Clustering (RecAgglo) for categorical data (2019)
- PRADA: Protection against DNN model stealing attacks (2019)
- Intel SGX related projects: Projects related to Intel Software Guard Extensions (SGX) (2018-2020)
- HardScope: Run-time scope enforcement on RISC-V (2018)
- MiniONN: Privacy-preserving neural networks (2017)
- C-FLAT: Control flow attestation for embedded systems software (2016)
- Open-TEE: Virtual TEE compliant with the GlobalPlatform TEE specification (2015)