Pushing Large Language Models to the 6G Edge: Vision, Challenges, and Opportunities
Bibliographic record
Abstract
Large language models (LLMs), which have shown remarkable capabilities, are revolutionizing artificial intelligence (AI) and shaping our society. However, the status quo cloud-based LLM deployment faces critical challenges such as long response time and privacy concerns, whereas on-device LLM deployment is hindered by the limited capabilities of end devices. To address the dilemma, this article explores the transformative potential of deploying LLMs at the 6G edge. We first introduce killer applications to exemplify the urgent need for edge LLM deployment and then, we identify the inherent limitations of on-device LLM deployment. We therefore argue that end-edge cooperation at the 6G edge is a promising solution for the dilemma. Towards this end, we elaborate on the 6G MEC architecture tailored for LLMs. Furthermore, we delve into edge training and edge inference for LLMs, with a focus on end-edge cooperation. In both aspects, we discuss a spectrum of cutting-edge techniques, including split learning/inference, parameter-efficient fine-tuning, parameter-sharing inference, and small-large language model cooperation. Finally, we investigate open problems in green and privacy-preserving edge LLM deployment. This work provides a comprehensive and forward-looking perspective and pathways for enabling LLM deployment at the network edge.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.005 | 0.016 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.001 | 0.002 |
| Scholarly communication | 0.004 | 0.011 |
| Open science | 0.003 | 0.005 |
| Research integrity | 0.003 | 0.007 |
| Insufficient payload (model declined to judge) | 0.003 | 0.003 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".