Federated Learning-Enabled Hybrid Language Models for Communication-Efficient Token Transmission

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Solat, Faranaksadat, Lee, Joohyung, Seif, Mohamed, Niyato, Dusit, Poor, H. Vincent
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909668052303872
author Solat, Faranaksadat
Lee, Joohyung
Seif, Mohamed
Niyato, Dusit
Poor, H. Vincent
author_facet Solat, Faranaksadat
Lee, Joohyung
Seif, Mohamed
Niyato, Dusit
Poor, H. Vincent
contents Hybrid Language Models (HLMs) combine the low-latency efficiency of Small Language Models (SLMs) on edge devices with the high accuracy of Large Language Models (LLMs) on centralized servers. Unlike traditional end-to-end LLM inference, HLMs reduce latency and communication by invoking LLMs only when local SLM predictions are uncertain, i.e., when token-level confidence is low or entropy is high. However, ambiguous or low-confidence predictions still require frequent offloading to the LLM, leading to significant communication overhead in bandwidth-constrained settings. To address this, we propose FedHLM, a communication-efficient HLM framework that integrates uncertainty-aware inference with Federated Learning (FL). FedHLM's key innovation lies in collaboratively learning token-level uncertainty thresholds that govern when LLM assistance is needed. Rather than using static or manually tuned thresholds, FedHLM employs FL to optimize these thresholds in a privacy-preserving, distributed manner. Additionally, it leverages embedding-based token representations for Peer-to-Peer (P2P) resolution, enabling clients to reuse tokens inferred by semantically similar peers without engaging the LLM. We further introduce hierarchical model aggregation: edge servers refine local routing policies through client updates, while cross-cluster coordination aligns global decision boundaries. This layered design captures recurring uncertainty patterns, reducing redundant LLM queries. Experiments on large-scale news classification tasks show that FedHLM reduces LLM transmissions by over 95 percent with negligible accuracy loss, making it well-suited for scalable and efficient edge-AI applications.
format Preprint
id arxiv_https___arxiv_org_abs_2507_00082
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Federated Learning-Enabled Hybrid Language Models for Communication-Efficient Token Transmission
Solat, Faranaksadat
Lee, Joohyung
Seif, Mohamed
Niyato, Dusit
Poor, H. Vincent
Machine Learning
Artificial Intelligence
Computation and Language
Hybrid Language Models (HLMs) combine the low-latency efficiency of Small Language Models (SLMs) on edge devices with the high accuracy of Large Language Models (LLMs) on centralized servers. Unlike traditional end-to-end LLM inference, HLMs reduce latency and communication by invoking LLMs only when local SLM predictions are uncertain, i.e., when token-level confidence is low or entropy is high. However, ambiguous or low-confidence predictions still require frequent offloading to the LLM, leading to significant communication overhead in bandwidth-constrained settings. To address this, we propose FedHLM, a communication-efficient HLM framework that integrates uncertainty-aware inference with Federated Learning (FL). FedHLM's key innovation lies in collaboratively learning token-level uncertainty thresholds that govern when LLM assistance is needed. Rather than using static or manually tuned thresholds, FedHLM employs FL to optimize these thresholds in a privacy-preserving, distributed manner. Additionally, it leverages embedding-based token representations for Peer-to-Peer (P2P) resolution, enabling clients to reuse tokens inferred by semantically similar peers without engaging the LLM. We further introduce hierarchical model aggregation: edge servers refine local routing policies through client updates, while cross-cluster coordination aligns global decision boundaries. This layered design captures recurring uncertainty patterns, reducing redundant LLM queries. Experiments on large-scale news classification tasks show that FedHLM reduces LLM transmissions by over 95 percent with negligible accuracy loss, making it well-suited for scalable and efficient edge-AI applications.
title Federated Learning-Enabled Hybrid Language Models for Communication-Efficient Token Transmission
topic Machine Learning
Artificial Intelligence
Computation and Language
url https://arxiv.org/abs/2507.00082