The following slides contain highlights of recent scientific achievements:
- May 22, 2026 — IntraShuffler: A Privacy Preserving Framework for Heterogeneous Differential Privacy (HDP)-Federated Learning (FL)
- May 22, 2026 — FedQueue: Queue-Aware Federated Learning for Cross-Facility HPC Training
- Apr 20, 2026 — XMark: Reliable Multi-Bit Watermarking for LLM-Generated Texts
- Apr 20, 2026 — SelfGrader: Stable Jailbreak Detection for Large Language Models using Token-Level Logits
- Apr 20, 2026 — Scalable Federated Learning for Scientific Foundation Models on Leadership-Class Systems
- Apr 20, 2026 — DP-TwoLevel: Two-Stage Gradient Subspace Learning for Differentially Private Federated Learning
- Apr 20, 2026 — Automated Membership Inference Attacks (MIA): Discovering MIA Signal Computations using Large Language Model (LLM) Agents
- Mar 17, 2026 — Differentially Private Federated Averaging with James-Stein Estimator
- Mar 17, 2026 — Energy–Performance Trade-offs in Privacy-Preserving Federated Learning on SmartNIC-Enabled HPC Systems
- Mar 17, 2026 — Selective Amnesia using Contrastive Subnet Erasure for Class Level Unlearning in Vision Models
- Mar 12, 2026 — FIRM: Federated Image Reconstruction Using Multimodal Tomographic Data
- Nov 18, 2025 — First Cross-Facility Federated Learning Deployment on DOE Leadership-Class Supercomputers
- Mar 24, 2025 — GridFM: Enabling Secure, Collaborative AI for Grid
- Mar 24, 2025 — FedSpaLLM: Federated Pruning of Large Language Models
- Mar 24, 2025 — Extending APPFL: Supporting Vertical, Hierarchical, and Decentralized FL for Science