Communication-Efficient Hybrid Language Model via Uncertainty-Aware Opportunistic and Compressed Transmission

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Oh, Seungeun, Kim, Jinhyuk, Park, Jihong, Ko, Seung-Woo, Choi, Jinho, Quek, Tony Q. S., Kim, Seong-Lyun
Format: Preprint
Veröffentlicht: 2025
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866913160659730432
author Oh, Seungeun
Kim, Jinhyuk
Park, Jihong
Ko, Seung-Woo
Choi, Jinho
Quek, Tony Q. S.
Kim, Seong-Lyun
author_facet Oh, Seungeun
Kim, Jinhyuk
Park, Jihong
Ko, Seung-Woo
Choi, Jinho
Quek, Tony Q. S.
Kim, Seong-Lyun
contents To support emerging language-based applications using dispersed and heterogeneous computing resources, the hybrid language model (HLM) offers a promising architecture, where an on-device small language model (SLM) generates draft tokens that are validated and corrected by a remote large language model (LLM). However, the original HLM suffers from substantial communication overhead, as the LLM requires the SLM to upload the full vocabulary distribution for each token. Moreover, both communication and computation resources are wasted when the LLM validates tokens that are highly likely to be accepted. To overcome these limitations, we propose communication-efficient and uncertainty-aware HLM (CU-HLM). In CU-HLM, the SLM transmits truncated vocabulary distributions only when its output uncertainty is high. We validate the feasibility of this opportunistic transmission by discovering a strong correlation between SLM's uncertainty and LLM's rejection probability. Furthermore, we theoretically derive optimal uncertainty thresholds and optimal vocabulary truncation strategies. Simulation results show that, compared to standard HLM, CU-HLM achieves up to 206$\times$ higher token throughput by skipping 74.8% transmissions with 97.4% vocabulary compression, while maintaining 97.4% accuracy.
format Preprint
id arxiv_https___arxiv_org_abs_2505_11788
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Communication-Efficient Hybrid Language Model via Uncertainty-Aware Opportunistic and Compressed Transmission
Oh, Seungeun
Kim, Jinhyuk
Park, Jihong
Ko, Seung-Woo
Choi, Jinho
Quek, Tony Q. S.
Kim, Seong-Lyun
Distributed, Parallel, and Cluster Computing
Information Theory
Machine Learning
Networking and Internet Architecture
Signal Processing
To support emerging language-based applications using dispersed and heterogeneous computing resources, the hybrid language model (HLM) offers a promising architecture, where an on-device small language model (SLM) generates draft tokens that are validated and corrected by a remote large language model (LLM). However, the original HLM suffers from substantial communication overhead, as the LLM requires the SLM to upload the full vocabulary distribution for each token. Moreover, both communication and computation resources are wasted when the LLM validates tokens that are highly likely to be accepted. To overcome these limitations, we propose communication-efficient and uncertainty-aware HLM (CU-HLM). In CU-HLM, the SLM transmits truncated vocabulary distributions only when its output uncertainty is high. We validate the feasibility of this opportunistic transmission by discovering a strong correlation between SLM's uncertainty and LLM's rejection probability. Furthermore, we theoretically derive optimal uncertainty thresholds and optimal vocabulary truncation strategies. Simulation results show that, compared to standard HLM, CU-HLM achieves up to 206$\times$ higher token throughput by skipping 74.8% transmissions with 97.4% vocabulary compression, while maintaining 97.4% accuracy.
title Communication-Efficient Hybrid Language Model via Uncertainty-Aware Opportunistic and Compressed Transmission
topic Distributed, Parallel, and Cluster Computing
Information Theory
Machine Learning
Networking and Internet Architecture
Signal Processing
url https://arxiv.org/abs/2505.11788