CoChat

Design and Implementation of a Lightweight Neural Controller for LLM Agent Systems

Open access
Brian Barnabas Langay · 2026 · International Journal of Latest Technology in Engineering Management & Applied Science

The DOI matches its Crossref record; 28 of the 28 references it cites were found in a registry or an index.

Abstract

Large language model (LLM) agents typically consume thousands of tokens on scaffolding system prompts, tool schemas, and conversation history before producing a single useful word. NEXUS (Neural EXecution & Understanding Substrate) is a 6.29-million-parameter neural controller that sits between a frozen LLM and its execution environment, replacing that token overhead with compact vector signals. The controller has five subsystems that run together (1) Protocol Cortex, which writes task descriptions directly into the LLM’s key-value cache so the model behaves as though it received detailed instructions without those instructions; (2) Belief Engine, a recurrent state-space model that tracks what the agent currently believes about its environment; (3) Resource Router, tool-selection classifier that uses explicit state-machine logic to guarantee valid tool calls; (4) Drift Sentinel, a lightweight monitor that detects when the agent’s output begins drifting off-task; and (5) Adapter Switch, which selects among small, low-rank weight updates (LoRA adapters) to specialize the LLM for different sub-tasks on the fly. We make three separate claims. First, we describe the architecture and the training recipe for all five components. Second, we report a deployment result: an open-source Model Context Protocol (MCP) server, nexus-mcp-oss, which achieves 72.86% fewer tokens delivered to the LLM in production through heuristic text-level compression (distinct from the KV-cache injection mechanism). Third, we present a controlled evaluation of the KV-cache injection mechanism itself, in which the Protocol Cortex is trained end-to-end with a frozen TinyLlama 1.1B and reaches a held-out perplexity of 8.91 versus 26,607 for an untrained baseline 2,987-fold improvement and 30–77 times lower perplexity than Prefix-Tuning, ActAdd, and LLMLingua at matched compression. Three of the four trained components converge on the synthetic benchmark; trained checkpoints, and benchmark data are publicly available. We are explicit about scope: the gains reported here are measured as token efficiency and predictive (perplexity) quality, and we set out in Sections 8.5 and 9.5 a concrete plan to test whether these efficiency gains carry through to downstream task quality: reasoning, planning, coding assistance, and multi-agent coordination.

Overview

Integrity signals

Registry

Crossref · 10.51583/ijltemas.2026.150500172

Cites

28 of 28 references verified0 citation errors · 0 not checkable

Statistics

Recomputing p-values needs the paper’s full text.

Verification

Claims the paper makes

The text we found for this paper looks like another document, so its claims and self-check weren’t run. The registry and citation checks still apply.

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

The text we found for this paper looks like another document, so the self-check wasn’t run.

Citations

Cites

Show all 28 references

Research works printed but not linked: 5

  1. Lewis, P., Perez, E., Piktus, A., Petroni, F., Karpukhin, V., Goyal, N., Küttler, H., Lewis, M., Yih, W., Rocktäschel, T., Riedel, S., & Kiela, D. (2020). Retrieval-augmented generation for knowledge-intensive NLP tasks. NeurIPS 2020.
  2. Hu, E. J., Shen, Y., Wallis, P., Allen-Zhu, Z., Li, Y., Wang, S., Wang, L., & Chen, W. (2021). LoRA: Low-rank adaptation of large language models. ICLR 2022.
  3. Vaswani, A., Shazeer, N., Parmar, N., Uszkoreit, J., Jones, L., Gomez, A. N., Kaiser, L., & Polosukhin, I. (2017). Attention is all you need. NeurIPS 2017.
  4. A. Gu, T. Dao. “Mamba: Linear-Time Sequence Modeling with Selective State Spaces.” arXiv:2312.00752, 2023. · arXiv
  5. Langay, B. B. (2026). Agent Workspace: Browser-Based Multi-Tool Integration and Sidecar Control for Autonomous LLM Agents. Master of Computer Science Thesis, Anhui University of Technology, Ma'anshan, China.

Metadata

Authors
Langay, B. B.
Corresponding author
Brian Barnabas Langay
Venue
International Journal of Latest Technology in Engineering Management & Applied Science
Published
2026-06-11
Volume
15
Issue
5
Pages
2138-2170
ISSN
2278-2540 (ISSN-L)
OpenAlex
W7164357331
Cited by
0
References
6
FWCI
0.00
Type
journal article
Language
en
Open access
Locations
International Journal of Latest Technology in Engineering Management & Applied Science (publishedVersion, cc-by)
Topics
  • Physical Sciences › Computer Science › Artificial Intelligence › Topic Modeling
  • Physical Sciences › Computer Science › Artificial Intelligence › Natural Language Processing Techniques
  • Physical Sciences › Computer Science › Artificial Intelligence › Artificial Intelligence Applications
Keywords
LLM agents, neural network controller, token efficiency, key-value cache, tool selection, state-space model, belief tracking, drift detection, LoRA adapters, low-rank adapters, Model Context Protocol, NEXUS

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

A second model labelled each claim on 2026-10-12 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review

How to cite

Cite CoChat for the checks on this page, and the paper for its own findings.

This page
CoChat. (2026). Citation check: Design and Implementation of a Lightweight Neural Controller for LLM Agent Systems. https://app.cochat.ai/papers/10.51583/ijltemas.2026.150500172
The paper
Langay, B. B. (2026). Design and Implementation of a Lightweight Neural Controller for LLM Agent Systems. International Journal of Latest Technology in Engineering Management & Applied Science. https://doi.org/10.51583/ijltemas.2026.150500172

Data sources

Bibliographic data: Semantic Scholar, OpenAlex, Crossref.

Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).

Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.