Neural Router: Semantic Content Matching for Agentic AI
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866917531528200192 |
|---|---|
| author | Lovén, Lauri Kumar, Abhishek Engelhardt, Alexander Saleh, Alaa Morabito, Roberto Liu, Xiaoli Motlagh, Naser Hossein Tarkoma, Sasu |
| author_facet | Lovén, Lauri Kumar, Abhishek Engelhardt, Alexander Saleh, Alaa Morabito, Roberto Liu, Xiaoli Motlagh, Naser Hossein Tarkoma, Sasu |
| contents | Large language models (LLMs) can serve as the semantic-matching engine of a content-based publish/subscribe broker for agentic AI across the edge-cloud computing continuum, bridging the vocabulary and modality gaps that defeat keyword and embedding filters. Framed as offline multi-label retrieval over three public datasets spanning social-media, legal, and smart-home sensor domains (six LLMs, seven baselines), our central contribution is a two-crossover cost-accuracy characterisation: an analytical context-window crossover below which a CoverAndMerge compression pipeline reduces LLM invocations, and an empirical discrimination-capacity crossover above which matching accuracy collapses independently of context budget, by a model-dependent factor of parameter count and training generation. Two findings carry practical weight: above the discrimination crossover, compression cannot recover accuracy and only frontier-scale models clear large subscription sets; and there backend choice dominates configuration choice, so model selection, not pipeline tuning, is the primary operator lever. We accompany this with three composable algorithms and a per-cluster Quality-of-Experience framework for autonomic LLM-tier selection. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2605_25701 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Neural Router: Semantic Content Matching for Agentic AI Lovén, Lauri Kumar, Abhishek Engelhardt, Alexander Saleh, Alaa Morabito, Roberto Liu, Xiaoli Motlagh, Naser Hossein Tarkoma, Sasu Distributed, Parallel, and Cluster Computing Computation and Language Information Retrieval Networking and Internet Architecture C.2.4; H.3.3; I.2.7 Large language models (LLMs) can serve as the semantic-matching engine of a content-based publish/subscribe broker for agentic AI across the edge-cloud computing continuum, bridging the vocabulary and modality gaps that defeat keyword and embedding filters. Framed as offline multi-label retrieval over three public datasets spanning social-media, legal, and smart-home sensor domains (six LLMs, seven baselines), our central contribution is a two-crossover cost-accuracy characterisation: an analytical context-window crossover below which a CoverAndMerge compression pipeline reduces LLM invocations, and an empirical discrimination-capacity crossover above which matching accuracy collapses independently of context budget, by a model-dependent factor of parameter count and training generation. Two findings carry practical weight: above the discrimination crossover, compression cannot recover accuracy and only frontier-scale models clear large subscription sets; and there backend choice dominates configuration choice, so model selection, not pipeline tuning, is the primary operator lever. We accompany this with three composable algorithms and a per-cluster Quality-of-Experience framework for autonomic LLM-tier selection. |
| title | Neural Router: Semantic Content Matching for Agentic AI |
| topic | Distributed, Parallel, and Cluster Computing Computation and Language Information Retrieval Networking and Internet Architecture C.2.4; H.3.3; I.2.7 |
| url | https://arxiv.org/abs/2605.25701 |