Large Language Models in Intent-Based Networking: a Comprehensive Survey Across the Intent Lifecycle
Bibliographic record
Abstract
The rising complexity of modern networks, particularly in 6 G environments, demands scalable and autonomous management frameworks. Intent-Based Networking (IBN) addresses this challenge by enabling users to specify high-level operational goals rather than low-level configurations. However, traditional IBN approaches remain limited by their reliance on strict intent interpretation mechanisms. Large Language Models (LLMs), with their advanced semantic understanding and contextual reasoning capabilities, can offer a promising enhancement to the IBN lifecycle. Hence, in this survey, we present the first dedicated and structured analysis of how LLMs are being integrated into the IBN paradigm. We examine the most recent literature to trace the application of LLMs across all five phases of the IBN lifecycle: intent profiling, translation, conflict resolution, policy activation, and assurance. Unlike prior works that treat LLMs and network management in isolation, this survey emphasizes their convergence, detailing how LLMs support context-aware interpretation, flexible policy generation, and dynamic adaptation in response to network variability. Additionally, we present a comprehensive taxonomy that maps current research efforts of LLM to IBN phases and the specific LLM models used in each phase. Furthermore, the survey offers an analysis of the limitations and open challenges associated with deploying LLMs into IBN systems.
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 | 0.004 | 0.014 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.002 | 0.002 |
| Bibliometrics | 0.004 | 0.005 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.006 | 0.010 |
| Open science | 0.003 | 0.002 |
| Research integrity | 0.002 | 0.004 |
| Insufficient payload (model declined to judge) | 0.005 | 0.002 |
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".