Structure-Aware NL-to-SQL for SFC Provisioning via AST-Masking Empowered Language Models
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2026
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866908785528799232 |
|---|---|
| author | Zhu, Xinyu Moshiri, Parisa Fard Lohan, Poonam Kantarci, Burak Janulewicz, Emil |
| author_facet | Zhu, Xinyu Moshiri, Parisa Fard Lohan, Poonam Kantarci, Burak Janulewicz, Emil |
| contents | Effective Service Function Chain (SFC) provisioning requires precise orchestration in dynamic and latency-sensitive networks. Reinforcement Learning (RL) improves adaptability but often ignores structured domain knowledge, which limits generalization and interpretability. Large Language Models (LLMs) address this gap by translating natural language (NL) specifications into executable Structured Query Language (SQL) commands for specification-driven SFC management. Conventional fine-tuning, however, can cause syntactic inconsistencies and produce inefficient queries. To overcome this, we introduce Abstract Syntax Tree (AST)-Masking, a structure-aware fine-tuning method that uses SQL ASTs to assign weights to key components and enforce syntax-aware learning without adding inference overhead. Experiments show that AST-Masking significantly improves SQL generation accuracy across multiple language models. FLAN-T5 reaches an Execution Accuracy (EA) of 99.6%, while Gemma achieves the largest absolute gain from 7.5% to 72.0%. These results confirm the effectiveness of structure-aware fine-tuning in ensuring syntactically correct and efficient SQL generation for interpretable SFC orchestration. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2601_17295 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Structure-Aware NL-to-SQL for SFC Provisioning via AST-Masking Empowered Language Models Zhu, Xinyu Moshiri, Parisa Fard Lohan, Poonam Kantarci, Burak Janulewicz, Emil Networking and Internet Architecture Computation and Language Machine Learning Effective Service Function Chain (SFC) provisioning requires precise orchestration in dynamic and latency-sensitive networks. Reinforcement Learning (RL) improves adaptability but often ignores structured domain knowledge, which limits generalization and interpretability. Large Language Models (LLMs) address this gap by translating natural language (NL) specifications into executable Structured Query Language (SQL) commands for specification-driven SFC management. Conventional fine-tuning, however, can cause syntactic inconsistencies and produce inefficient queries. To overcome this, we introduce Abstract Syntax Tree (AST)-Masking, a structure-aware fine-tuning method that uses SQL ASTs to assign weights to key components and enforce syntax-aware learning without adding inference overhead. Experiments show that AST-Masking significantly improves SQL generation accuracy across multiple language models. FLAN-T5 reaches an Execution Accuracy (EA) of 99.6%, while Gemma achieves the largest absolute gain from 7.5% to 72.0%. These results confirm the effectiveness of structure-aware fine-tuning in ensuring syntactically correct and efficient SQL generation for interpretable SFC orchestration. |
| title | Structure-Aware NL-to-SQL for SFC Provisioning via AST-Masking Empowered Language Models |
| topic | Networking and Internet Architecture Computation and Language Machine Learning |
| url | https://arxiv.org/abs/2601.17295 |