Benchgen
4 publications

Research

BenchGen is built on original research into how AI agents plan, reason, and operate over long horizons. Below are our peer-reviewed publications on agent architectures, multi-agent coordination, and LLM-based systems — the foundations behind the BenchGen platform.

First page preview of "Thought Management System for Long-Horizon, Goal-Driven LLM Agents"
2026

Thought Management System for Long-Horizon, Goal-Driven LLM Agents

Journal of Computational Science · Elsevier · Vol. 93 · Article 102740

DOI ScienceDirect
First page preview of "Optimizing Data Exchange in Multi-Agent Systems via Execution Environment Variables"
2025

Optimizing Data Exchange in Multi-Agent Systems via Execution Environment Variables

Nauka ta Tekhnika Sohodni · No. 10(51) · pp. 1250–1265

DOI Full text
First page preview of "LLMAgentNet: A Collaborative Network of Autonomous AI Agents for Complex Task Execution"
2024

LLMAgentNet: A Collaborative Network of Autonomous AI Agents for Complex Task Execution

Scientific Bulletin of UNFU Ukraine · No. 4(35) · pp. 101–104

DOI Full text
First page preview of "Using Large Language Models for Domain-Specific Language Code Generation"
2023

Using Large Language Models for Domain-Specific Language Code Generation

Herald of Khmelnytskyi National University · No. 4 · pp. 21–24

DOI Full text