FedReason: Large Language Models as Weight-Space Meta-Aggregators for Heterogeneous Federated Learning

Summary

FedReason uses a large language model to choose the server-side aggregation rule in federated learning, calling it every ten rounds to analyse client weights in a restricted sandbox and submit a short Python aggregation program, falling back to FedAvg when output is invalid. Evaluated with Claude Opus 5.5 on CIFAR-10 and CIFAR-100 with ResNet-18 across 168 runs, it beat FedAvg by 1.26 percentage points on average across 12 paired label-skew runs. It remained about one point below FLTrust under Byzantine clients, and accuracy fell from 73.8% to 38.8% in one seed without proxy feedback.

Related items

• InfoPerplexity launches its own take on AI agents for workSame vendor · The Verge (AI)
• MediumFake OpenAI, SpaceX events used to steal Microsoft, Google accountsSame vendor · BleepingComputer
• InfoClaude AI can plug directly into your bank accountSame vendor · The Verge (AI)
• InfoReview of the "Risks from automated R&D" section in Anthropic's Risk Report (August 2026)Same vendor · METR
• InfoIndependent Assessment of AI R&D Acceleration at Anthropic in October 2026Same vendor · METR