FERA: Uncertainty-Aware Federated Reasoning for Large Language Models

• URL: http://arxiv.org/abs/2605.10082v1
• Date: Mon, 11 May 2026 07:04:51 GMT
• Title: FERA: Uncertainty-Aware Federated Reasoning for Large Language Models
• Abstract summary: We study federated reasoning, where a server improves multi-step reasoning by coordinating with heterogeneous clients holding private demonstrations.<n>The key challenge is that client reliability is query-dependent, while the server cannot inspect client data to determine which contributions are trustworthy.<n>We propose Uncertainty-Aware Federated Reasoning (FERA), a training-free framework based on iterative server-client co-refinement.
• Score: 60.52562148874846
• License: http://creativecommons.org/licenses/by-nc-sa/4.0/
• Abstract: Large language models (LLMs) exhibit strong reasoning capabilities when guided by high-quality demonstrations, yet such data is often distributed across organizations that cannot centralize it due to regulatory, proprietary, or institutional constraints. We study federated reasoning, where a server improves multi-step reasoning by coordinating with heterogeneous clients holding private demonstrations, without centralized training or raw data sharing. The key challenge is that client reliability is query-dependent, while the server cannot inspect client data to determine which contributions are trustworthy. To address this, we propose Uncertainty-Aware Federated Reasoning (FERA), a training-free framework based on iterative server-client co-refinement. Across communication rounds, clients generate reasoning traces with lightweight uncertainty estimates, and the server synthesizes them into improved reasoning that is redistributed as context for the next round, progressively improving both server outputs and client-side reasoning. Within each round, Uncertainty-Aware Self-Critique Aggregation (UA-SCA) resolves conflicts among heterogeneous client traces through query-dependent trust weighting and structured cross-client verification. Rather than simply discarding low-quality traces, UA-SCA revises flawed reasoning steps to recover useful information. We provide theoretical guarantees showing that the proposed iterative protocol converges and that uncertainty-aware weighting accelerates convergence. Experiments on multiple reasoning benchmarks show that FERA consistently outperforms both federated training and training-free baselines, achieving progressively higher accuracy across rounds while maintaining communication and computational efficiency.

Related papers

• Confidence-Aware Alignment Makes Reasoning LLMs More Reliable [65.44962502963378]
  We introduce CASPO, a framework that aligns token-level confidence with step-wise logical correctness without training a separate reward model.<n>During inference, we propose Confidence-aware Thought (CaT), which dynamically prunes uncertain reasoning branches with negligible O(V) latency.<n>Experiments across ten benchmarks and multiple model families show that CASPO consistently improves reasoning reliability and inference efficiency.
  arXiv  Detail & Related papers  (2026-05-08T07:08:25Z)
• Byzantine-Robust and Differentially Private Federated Optimization under Weaker Assumptions [89.52532304099522]
  Federated Learning (FL) enables clients to collaboratively train a shared model without centralizing their raw data, offering an inherent level of privacy.<n> gradients and model updates can still leak sensitive information, while malicious servers may mount adversarial attacks such as Byzantine manipulation.<n>These vulnerabilities highlight the need to address differential privacy (DP) and Byzantine robustness within a unified framework.<n>We propose Byz-Clip21-SGD2M, a new algorithm that integrates robust aggregation with double momentum and carefully designed clipping.
  arXiv  Detail & Related papers  (2026-03-24T17:39:09Z)
• ERIS: Enhancing Privacy and Communication Efficiency in Serverless Federated Learning [6.486831630436399]
  ERIS is a serverless FL framework that balances privacy and accuracy while eliminating the server bottleneck and distributing the communication load.<n>We theoretically prove that ERIS converges at the same rate as FedAvg under standard assumptions, and (ii) bounds mutual information leakage inversely with the number of aggregators.
  arXiv  Detail & Related papers  (2026-02-09T13:05:41Z)
• Stragglers Can Contribute More: Uncertainty-Aware Distillation for Asynchronous Federated Learning [61.249748418757946]
  Asynchronous federated learning (FL) has recently gained attention for its enhanced efficiency and scalability.<n>We propose FedEcho, a novel framework that incorporates uncertainty-aware distillation to enhance the asynchronous FL performances.<n>We demonstrate that FedEcho consistently outperforms existing asynchronous federated learning baselines.
  arXiv  Detail & Related papers  (2025-11-25T06:25:25Z)
• FLARE: Adaptive Multi-Dimensional Reputation for Robust Client Reliability in Federated Learning [0.6524460254566904]
  Federated learning (FL) enables collaborative model training while preserving data privacy.<n>It remains vulnerable to malicious clients who compromise model integrity through Byzantine attacks, data poisoning, or adaptive adversarial behaviors.<n>We propose FLARE, an adaptive reputation-based framework that transforms client reliability assessment from binary decisions to a continuous, multi-dimensional trust evaluation.
  arXiv  Detail & Related papers  (2025-11-18T17:57:40Z)
• Byzantine-Robust Federated Learning Using Generative Adversarial Networks [1.4091801425319963]
  Federated learning (FL) enables collaborative model training across distributed clients without sharing raw data, but its robustness is threatened by Byzantine behaviors such as data and model poisoning.<n>We present a defense framework that addresses these challenges by leveraging a conditional generative adversarial network (cGAN) at the server to synthesize representative data for validating client updates.<n>This approach eliminates reliance on external datasets, adapts to diverse attack strategies, and integrates seamlessly into standard FL.
  arXiv  Detail & Related papers  (2025-03-26T18:00:56Z)
• Secure and Confidential Certificates of Online Fairness [8.681039636447865]
  Confidentiality is critical for delivering ML services legally and responsibly.<n>Existing methods that assess model fairness confidentially lack either (i) reliability because they certify fairness with respect to a static set of data.<n>We present OATH, a deployably efficient and scalable zero-knowledge proof protocol for confidential online group fairness certification.
  arXiv  Detail & Related papers  (2024-09-17T16:00:35Z)
• Certifiably Byzantine-Robust Federated Conformal Prediction [49.23374238798428]
  We introduce a novel framework Rob-FCP, which executes robust federated conformal prediction effectively countering malicious clients. We empirically demonstrate the robustness of Rob-FCP against diverse proportions of malicious clients under a variety of Byzantine attacks.
  arXiv  Detail & Related papers  (2024-06-04T04:43:30Z)
• Risk-Aware Accelerated Wireless Federated Learning with Heterogeneous Clients [21.104752782245257]
  Wireless Federated Learning (FL) is an emerging distributed machine learning paradigm. This paper proposes a novel risk-aware accelerated FL framework that accounts for the clients heterogeneity in the amount of possessed data. The proposed scheme is benchmarked against a conservative scheme (i.e., only allowing trustworthy devices) and an aggressive scheme (i.e. oblivious to the trust metric)
  arXiv  Detail & Related papers  (2024-01-17T15:15:52Z)
• FedSkip: Combatting Statistical Heterogeneity with Federated Skip Aggregation [95.85026305874824]
  We introduce a data-driven approach called FedSkip to improve the client optima by periodically skipping federated averaging and scattering local models to the cross devices. We conduct extensive experiments on a range of datasets to demonstrate that FedSkip achieves much higher accuracy, better aggregation efficiency and competing communication efficiency.
  arXiv  Detail & Related papers  (2022-12-14T13:57:01Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.