Pith. sign in

REVIEW 30 cited by

Pushing Large Language Models to the 6G Edge: Vision, Challenges, and Opportunities

Not yet reviewed by Pith; the record is open.

Re-run · record.json Download PDF Read on arXiv ↗

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2309.16739 v4 pith:7REVKYVF submitted 2023-09-28 cs.LG cs.AI

Zheng Lin , Guanqiao Qu , Qiyuan Chen , Xianhao Chen , Zhe Chen , Kaibin Huang This is my paper · ORCID
classification cs.LGcs.AI
keywords edgellmschallengesdeploymentinferencearticleaspectscritical
verification ladder T0 review T1 audit T2 compute T3 formal

Signed reviews

No signed human review yet.

0 comments
read the original abstract

Large language models (LLMs), which have shown remarkable capabilities, are revolutionizing AI development and potentially shaping our future. However, given their multimodality, the status quo cloud-based deployment faces some critical challenges: 1) long response time; 2) high bandwidth costs; and 3) the violation of data privacy. 6G mobile edge computing (MEC) systems may resolve these pressing issues. In this article, we explore the potential of deploying LLMs at the 6G edge. We start by introducing killer applications powered by multimodal LLMs, including robotics and healthcare, to highlight the need for deploying LLMs in the vicinity of end users. Then, we identify the critical challenges for LLM deployment at the edge and envision the 6G MEC architecture for LLMs. Furthermore, we delve into two design aspects, i.e., edge training and edge inference for LLMs. In both aspects, considering the inherent resource limitations at the edge, we discuss various cutting-edge techniques, including split learning/inference, parameter-efficient fine-tuning, quantization, and parameter-sharing inference, to facilitate the efficient deployment of LLMs. This article serves as a position paper for thoroughly identifying the motivation, challenges, and pathway for empowering LLMs at the 6G edge.

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 30 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. Dynamic Uncertainty-aware Multimodal Fusion for Outdoor Health Monitoring

    cs.NI 2025-08 unverdicted novelty 6.0 of 10

    DUAL-Health is an uncertainty-aware multimodal fusion framework that quantifies sensor noise, customizes fusion weights accordingly, and aligns modality distributions to improve outdoor health monitoring.

  2. RRTO: A High-Performance Transparent Offloading System for Model Inference in Mobile Edge Computing

    cs.NI 2025-07 conditional novelty 6.0 of 10

    RRTO identifies static inference operator sequences from CUDA call logs alone and replays them on an edge GPU, cutting transparent-offloading communication to 11 RPCs per inference instead of thousands, with performan...

  3. Fast and Cost-effective Speculative Edge-Cloud Decoding with Early Exits

    cs.RO 2025-05 conditional novelty 6.0 of 10

    Edge-cloud speculative decoding runs faster when early exits in the server model let the client pre-draft the next candidate tokens before final verification is complete.

  4. WAFBOOSTER: Automatic Boosting of WAF Security Against Mutated Malicious Payloads

    cs.CR 2025-01 reject novelty 6.0 of 10

    WAFBOOSTER combines a shadow model, an RNN payload generator, and automatic signature extraction to harden web application firewalls, but its headline rejection-rate improvement is measured on the same payloads used t...

  5. Hierarchical Split Federated Learning: Convergence Analysis and System Optimization

    cs.LG 2024-12 conditional novelty 6.0 of 10

    Derives a convergence bound for hierarchical split federated learning and uses it to jointly optimize model-splitting points and per-tier aggregation intervals.

  6. AIC-VDS: Attention-Based In-Context Learning for Joint Velocity Control and Data Collection Scheduling in Multi-UAV-Assisted Pipeline Monitoring

    cs.AI 2025-10 reject novelty 5.0 of 10

    AIC-VDS uses trainable attention to shrink sensor data prompts for an LLM, and simulations show lower packet loss than two baselines in multi-UAV monitoring.

  7. PHandover: Parallel Handover in Mobile Satellite Network

    cs.NI 2025-07 conditional novelty 5.0 of 10

    A parallel, plan-based handover using a new Satellite Synchronized Function cuts LEO satellite handover latency to about 9 ms on average in an emulated prototype.

  8. Prompting Wireless Networks: Reinforced In-Context Learning for Power Control

    eess.SP 2025-06 conditional novelty 5.0 of 10

    Prompting LLMs with a few reward-ranked state-action examples controls base station power at a level comparable to a trained DQN on a small simulated problem.

  9. Recursive Offloading for LLM Serving in Multi-tier Networks

    cs.DC 2025-05 conditional novelty 5.0 of 10

    RecServe routes LLM requests across device, edge, and cloud using adaptive confidence thresholds based on recent history, cutting communication by over 50% versus cloud-only serving while keeping quality close to cloud-level.

  10. Serving Long-Context LLMs at the Mobile Edge: Test-Time Reinforcement Learning-based Model Caching and Inference Offloading

    cs.NI 2025-01 reject novelty 5.0 of 10

    A test-time reinforcement learning framework for joint model caching and inference offloading is claimed to cut simulated long-context LLM serving costs at the mobile edge by at least 30%.

  11. Rethinking Membership Inference Attacks Against Transfer Learning

    cs.CR 2025-01 conditional novelty 5.0 of 10

    A white-box attack on the student model can infer teacher-training membership in transfer learning by comparing the student's hidden representations with those of a shadow student model.

  12. Constructing 4D Radio Map in LEO Satellite Networks with Limited Samples

    cs.NI 2025-01 conditional novelty 5.0 of 10

    DeepRM uses neural networks to solve compressive sensing and tensor decomposition, reconstructing 4D radio maps (space plus frequency) with fewer samples and sensors than classical baselines.

  13. LEO-Split: A Semi-Supervised Split Learning Framework over LEO Satellite Networks

    cs.LG 2025-01 conditional novelty 5.0 of 10

    A semi-supervised split learning framework with an auxiliary client model, adaptive pseudo-label thresholds, and activation interpolation improves training speed and accuracy over LEO satellite links.

  14. CP-Guard: Malicious Agent Detection and Defense in Collaborative Bird's Eye View Perception

    cs.AI 2024-12 conditional novelty 4.5 of 10

    A collaborative perception defense that uses recursive group consensus checks and a consistency loss to filter malicious agents, without needing prior attack probabilities.

  15. Joint Partitioning and Placement of Foundation Models for Real-Time Edge AI

    cs.DC 2025-11 reject novelty 4.0 of 10

    A framework for runtime re-splitting and re-placement of foundation model layers across edge nodes is proposed, but its claimed latency gains are inherited from prior work rather than measured.

  16. Energy-Efficient Wireless LLM Inference via Uncertainty and Importance-Aware Speculative Decoding

    cs.LG 2025-08 reject novelty 4.0 of 10

    This paper proposes filtering cloud-verification requests by combining token-level uncertainty with attention-based importance, claiming energy savings up to 40.7% in wireless hybrid LLM inference.

  17. Toward Edge General Intelligence with Multiple-Large Language Model (Multi-LLM): Architecture, Trust, and Orchestration

    cs.NI 2025-07 conditional novelty 4.0 of 10

    A survey of multi-LLM systems in edge computing, covering architectures, enabling technologies, trust mechanisms, applications, and open datasets for edge general intelligence.

  18. SFL-LEO: Asynchronous Split-Federated Learning Design for LEO Satellite-Ground Network Framework

    cs.NI 2025-04 reject novelty 4.0 of 10

    An asynchronous split-federated learning framework for LEO satellite-ground networks enables local updates during disconnection and heterogeneous model splitting, with simulated accuracy gains over split learning and ...

  19. Multimodal-to-Text Prompt Engineering in Large Language Models Using Feature Embeddings for GNSS Interference Characterization

    cs.AI 2025-01 reject novelty 4.0 of 10

    Retrieval-augmented LLaVA is applied to GNSS interference classification, but the reported accuracy is unvalidated because query prompts contain the true labels.

  20. LCFed: An Efficient Clustered Federated Learning Framework for Heterogeneous Data

    cs.LG 2025-01 conditional novelty 4.0 of 10

    LCFed combines model splitting with clustered federated learning to share global and cluster-level knowledge, and uses low-rank model projections to cut clustering cost.

  21. From Connectivity to Autonomy: The Dawn of Self-Evolving Communication Systems

    eess.SY 2025-05 conditional novelty 3.0 of 10

    The paper sketches a four-layer AI-enabled architecture for self-evolving 6G networks and a roadmap to implement it.

  22. Secure Resource Allocation via Constrained Deep Reinforcement Learning

    cs.LG 2025-01 reject novelty 3.0 of 10

    A deep Q-network with a fixed deadline penalty is claimed to cut simulated system cost by up to 40% and energy use by 41.5% in serverless multi-cloud offloading.

  23. From Large AI Models to Agentic AI: A Tutorial on Future Intelligent Communications

    cs.AI 2025-05 conditional novelty 2.0 of 10

    This paper is a broad tutorial on applying LAMs and agentic AI to 6G, largely restating existing research rather than introducing new results.

  24. The Future of Internet of Things and Multimodal Language Models in 6G Networks: Opportunities and Challenges

    cs.CY 2025-04 conditional novelty 2.0 of 10

    A narrative survey arguing that combining IoT, multimodal language models, and 6G can improve smart applications, with a taxonomy of sensors, communication, processing, and security.

  25. Vision-Language Models for Edge Networks: A Comprehensive Survey

    cs.CV 2025-02 reject novelty 2.0 of 10

    A survey of lightweight vision-language models for edge deployment, marred by citation errors, self-citation, and a lack of selection methodology.

  26. A Survey on Backdoor Threats in Large Language Models (LLMs): Attacks, Defenses, and Evaluations

    cs.CR 2025-02 conditional novelty 2.0 of 10

    A literature review that taxonomizes LLM backdoor attacks and defenses by model construction phase, with no new experimental results.

  27. Integrating LLMs with ITS: Recent Advances, Potentials, Challenges, and Future Directions

    eess.SY 2025-01 conditional novelty 2.0 of 10

    The paper surveys recent work, models, applications, and challenges of using LLMs in intelligent transportation systems, without presenting new experimental results.

  28. A Comprehensive Survey of Large AI Models for Future Communications: Foundations, Applications and Challenges

    cs.IT 2025-05 conditional novelty 1.0 of 10

    A survey organizing the growing literature on large AI models for 6G communications, with a classification of model types, training and evaluation methods, and a list of challenges.

  29. A Contemporary Survey of Large Language Model Assisted Program Analysis

    cs.SE 2025-02 conditional novelty 1.0 of 10

    A review that catalogs how large language models are used in static, dynamic, and hybrid program analysis, and outlines open challenges.

  30. White paper: Towards Human-centric and Sustainable 6G Services -- the fortiss Research Perspective

    cs.NI 2025-07 unverdicted

    A research institute's white paper restating known 6G trends; no new technical results or measurements are presented.

Pith tools