HKUST's AuditFL adds a blockchain clearing layer for auditable federated learning
HKUST · hf · 2026-11-03
AuditFL from HKUST introduces a contract-backed on-chain clearing layer for open federated learning, making rounds auditable and economically constrained without removing the aggregator. Only identifiers and economic state hit the chain while tensors stay off-chain; clients post deposits and earn refunds/rewards, preserving model accuracy under failures.
More from Research
- Scalable gradient attribution finds deceptively harmful training data, with a 0.6B model as monitor — PeterHndrsn · 2026-11-03
- Dynamical releases SDL-2: frontier agents make correct materials-science calls in only ~half of experiments — BenBlaiszik · 2026-11-03
- Follow-up: Astra's weight edits only worked on Qwen's visible flaw, not on Gemma 270m — teortaxesTex · 2026-11-03
- Zuckerberg's Biohub bets $1.8B on virtual cells: the bottleneck is data, not AI — HarperSCarroll · 2026-11-03
- EMBER: pretrained gene-centric foundation model predicts multigene perturbation effects — jmuiuc · 2026-11-03
- New preprint RLCE: reinforcement learning with comparative evidence for social intelligence — pliang279 · 2026-11-03