Jenius Agent: Towards Experience-Driven Accuracy Optimization in Real-World Scenarios

TL;DR

This work addresses the reliability and efficiency of LLM-based autonomous agents operating in real-world environments. It presents Jenius-Agent, a modular framework that combines adaptive prompt generation, context-aware tool orchestration, and hierarchical memory management to improve task grounding, tool usage accuracy, and context retention through a closed-loop reasoning–action–memory cycle. The authors validate the approach with a dual-benchmark evaluation (APIGen for single-turn tool use and Jenius-bench for multi-turn dialogues) and demonstrate up to $20\%$ gains in task accuracy and substantial token-cost reductions, alongside high-quality, protocol-aligned real-world deployment. The results establish a practical blueprint for production-ready autonomous agents with robust performance and scalable memory, facilitated by MCP-based tooling and a disciplined evaluation framework.

Abstract

As agent systems powered by large language models (LLMs) advance, improving the task performance of an autonomous agent, especially in context understanding, tool usage, and response generation, has become increasingly critical. Although prior studies have advanced the overall design of LLM-based agents, systematic optimization of their internal reasoning and tool-use pipelines remains underexplored. This paper introduces an agent framework grounded in real-world practical experience, with three key innovations: (1) an adaptive prompt generation strategy that aligns with the agent's state and task goals to improve reliability and robustness; (2) a context-aware tool orchestration module that performs tool categorization, semantic retrieval, and adaptive invocation based on user intent and context; and (3) a layered memory mechanism that integrates session memory, task history, and external summaries to improve relevance and efficiency through dynamic summarization and compression. An end-to-end framework named Jenius-Agent has been integrated with three key optimizations, including tools based on the Model Context Protocol (MCP), file input/output (I/O), and execution feedback. The experiments show a 20 percent improvement in task accuracy, along with a reduced token cost, response latency, and invocation failures. The framework is already deployed in Jenius (https://www.jenius.cn), providing a lightweight and scalable solution for robust, protocol-compatible autonomous agents.

Figures (6)

• Figure 1: A typical ReAct style autonomous agent workflow.
• Figure 2: Jenius Agent Framework. The LLM acts as the central orchestrator, coordinating task execution with three core modules: adaptive prompt generation, tool retrieval, and memory management that enhance adaptability and efficiency.
• Figure 3: Adaptive Prompt Generation.
• Figure 4: Message Summarization Process.
• Figure 5: Comprehensive agent evaluation framework for Procedural, Semantic, and Efficiency Dimensions.