Federated Language Models Under Bandwidth Budgets: Distillation Rates and Conformal Coverage

• URL: http://arxiv.org/abs/2605.09986v1
• Date: Mon, 11 May 2026 05:01:43 GMT
• Title: Federated Language Models Under Bandwidth Budgets: Distillation Rates and Conformal Coverage
• Abstract summary: Training a language model on data scattered across bandwidth-limited nodes that cannot be centralized is a setting that arises in clinical networks, enterprise knowledge bases, and scientific consortia.<n>We study the regime in which data must remain distributed across nodes, and ask what statistical guarantees are in principle achievable under explicit bandwidth budgets.
• Score: 12.805268849262243
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Training a language model on data scattered across bandwidth-limited nodes that cannot be centralized is a setting that arises in clinical networks, enterprise knowledge bases, and scientific consortia. We study the regime in which data must remain distributed across nodes, and ask what statistical guarantees are in principle achievable under explicit bandwidth budgets; we aim to characterize what is provably possible, not to demonstrate a deployment-ready system. Existing theory treats either training-time consistency or inference-time calibration in isolation, and none makes bandwidth a first-class statistical parameter. We analyze two protocols, Federated Probe-Logit Distillation (FPLD) for training and Federated Conformal RAG (FC-RAG) for inference, as the analytical vehicles for our results. Our first main result is an explicit high-probability KL-consistency rate for FPLD with simultaneous dependence on node count $K$, per-node sample size $n$, quantization budget $B$, probe-set size $m$, and vocabulary size $V$; bandwidth enters only through an exponentially vanishing quantization term. Our second main result is a distribution-free marginal-coverage bound for FC-RAG, whose novel retrieval-bandwidth slack $Δ_{\mathrm{RAG}} = f_{\max}\sqrt{K^{-2}\sum_i v(B_i)}$ makes per-node retrieval bandwidth a first-class statistical parameter, with arithmetic aggregation across $K$ nodes shrinking the slack as $K^{-1/2}$ in the per-node-uniform regime. A Pinsker-type corollary composes the two bounds into an end-to-end coverage guarantee. Synthetic experiments verify the predicted scaling along the bounds' parameters; small-scale experiments on a GPT-2 testbed illustrate that the qualitative bandwidth-accuracy tradeoff survives on a real language model. A deployment-scale empirical evaluation is out of scope.

Related papers

• Tight Convergence Rates for Online Distributed Linear Estimation with Adversarial Measurements [66.94250413799232]
  We study mean estimation of a random vector $X$ in a distributed parameter-server-worker setup.<n>Key challenges are adversarial measurements and asynchrony.<n>Results provide a unified finite-time characterization of robustness, identifiability, and statistical efficiency in distributed linear estimation with adversarial workers.
  arXiv  Detail & Related papers  (2026-04-07T11:45:55Z)
• MDNS: Masked Diffusion Neural Sampler via Stochastic Optimal Control [48.504188275208556]
  We study the problem of learning a neural sampler to generate samples from discrete state spaces where the target probability mass function $piproptomathrme-U$ is known up to normalizing constant.<n>We propose $textbfM$asked $textbfDiffusion, a novel framework for discrete neural samplers by aligning measures through a family of learning objectives.
  arXiv  Detail & Related papers  (2025-08-14T14:27:16Z)
• Probabilistically Tightened Linear Relaxation-based Perturbation Analysis for Neural Network Verification [83.25968588249776]
  We present a novel framework that combines over-approximation techniques from LiRPA-based approaches with a sampling-based method to compute tight intermediate reachable sets.<n>With negligible computational overhead, $textttPT-LiRPA$ exploiting the estimated reachable sets, significantly tightens the lower and upper linear bounds of a neural network's output.
  arXiv  Detail & Related papers  (2025-07-07T18:45:53Z)
• Optimal and Feasible Contextuality-based Randomness Generation [4.2126604059714685]
  Semi-independent (device-independent) randomness generation protocols based on Kochen-Specker contextuality offer the attractive features of compact devices.<n>We show that a single qubit is non-contextual, in that there exist qubit correlations that cannot be explained by $epsilon$-faithful NCHV models.<n>We point out possible attacks by quantum and general consistent (non-signalling) adversaries for certain classes of contextuality tests.
  arXiv  Detail & Related papers  (2024-12-28T12:11:07Z)
• Certifiably Robust Model Evaluation in Federated Learning under Meta-Distributional Shifts [8.700087812420687]
  We provide guarantees for the model's performance on a different, unseen network "B"<n>We show how the principled vanilla DKW bound enables certification of the model's true performance on unseen clients within the same (source) network.
  arXiv  Detail & Related papers  (2024-10-26T18:45:15Z)
• Doubly Robust Conditional Independence Testing with Generative Neural Networks [8.323172773256449]
  This article addresses the problem of testing the conditional independence of two generic random vectors $X$ and $Y$ given a third random vector $Z$. We propose a new non-parametric testing procedure that avoids explicitly estimating any conditional distributions.
  arXiv  Detail & Related papers  (2024-07-25T01:28:59Z)
• Learning with Norm Constrained, Over-parameterized, Two-layer Neural Networks [54.177130905659155]
  Recent studies show that a reproducing kernel Hilbert space (RKHS) is not a suitable space to model functions by neural networks. In this paper, we study a suitable function space for over- parameterized two-layer neural networks with bounded norms.
  arXiv  Detail & Related papers  (2024-04-29T15:04:07Z)
• Collaborative non-parametric two-sample testing [55.98760097296213]
  The goal is to identify nodes where the null hypothesis $p_v = q_v$ should be rejected. We propose the non-parametric collaborative two-sample testing (CTST) framework that efficiently leverages the graph structure. Our methodology integrates elements from f-divergence estimation, Kernel Methods, and Multitask Learning.
  arXiv  Detail & Related papers  (2024-02-08T14:43:56Z)
• On the Convergence of Federated Averaging under Partial Participation for Over-parameterized Neural Networks [13.2844023993979]
  Federated learning (FL) is a widely employed distributed paradigm for collaboratively machine learning models from multiple clients without sharing local data. In this paper, we show that FedAvg converges to a global minimum at a global rate at a global focus.
  arXiv  Detail & Related papers  (2023-10-09T07:56:56Z)
• Self-Supervised Scalable Deep Compressed Sensing [24.854496459622787]
  Compressed sensing is a promising tool for reducing sampling costs. Current deep neural network (NN)-based CS methods face the challenges of collecting labeled measurement-ground truth (GT) data. This paper proposes a novel $mathbfS$elf-supervised s$mathbfC$alable deep CS method.
  arXiv  Detail & Related papers  (2023-08-26T06:03:06Z)
• On the Effective Number of Linear Regions in Shallow Univariate ReLU Networks: Convergence Guarantees and Implicit Bias [50.84569563188485]
  We show that gradient flow converges in direction when labels are determined by the sign of a target network with $r$ neurons. Our result may already hold for mild over- parameterization, where the width is $tildemathcalO(r)$ and independent of the sample size.
  arXiv  Detail & Related papers  (2022-05-18T16:57:10Z)
• The Fundamental Price of Secure Aggregation in Differentially Private Federated Learning [34.630300910399036]
  We characterize the fundamental communication cost required to obtain the best accuracy under $varepsilon$ central DP. Our results show that $tildeOleft( min(n2varepsilon2, d) right)$ bits per client are both sufficient and necessary. This provides a significant improvement relative to state-of-the-art SecAgg distributed DP schemes.
  arXiv  Detail & Related papers  (2022-03-07T22:56:09Z)

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.