FERA improves LLM multi-step reasoning in federated settings via iterative uncertainty-aware aggregation of client traces without data sharing or training.
Question:{query} Answer:Let’s think step by step.{text} Server Query Initialization Prompt for MMLU-Pro & AQUA-RAT Standard QA Benchmark You are taking a multiple-choice question
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
other 1
citation-polarity summary
fields
cs.CL 1years
2026 1verdicts
UNVERDICTED 1roles
other 1polarities
unclear 1representative citing papers
citing papers explorer
-
FERA: Uncertainty-Aware Federated Reasoning for Large Language Models
FERA improves LLM multi-step reasoning in federated settings via iterative uncertainty-aware aggregation of client traces without data sharing or training.