What question did this study set out to answer?

This work aims to develop a heterogeneous agentic AI system for efficient and context-aware UI generation using small language models.

May 16, 2026Open Access

Heterogeneous Agentic AI System Using Fine-Tuned SLMs

Puntos clave

This work aims to develop a heterogeneous agentic AI system for efficient and context-aware UI generation using small language models.
Developed a framework splitting orchestration and generation tasks across models.
Utilized Gemma 4 E2B for requirements elicitation and Qwendean 4B for UI code generation, integrating datasets from TailwindCSS and ShadCN.
Implemented fine-tuning using LoRA and rsLoRA techniques, with iterative dataset augmentation increasing samples from 900 to 4,100.
Demonstrated a 3-5× reduction in latency in local section-based UI generation.
Showed significant improvements in output quality consistency.
Successfully synthesized coherent interfaces with React/TypeScript code while integrating TailwindCSS components.

Resumen

In this work, a heterogeneous agentic AI framework is proposed for context-aware UI generation, which makes use of fine-tuned small language models (SLMs). In particular, the proposed framework splits tasks of orchestration and domain-specific generation between models, where Gemma 4 E2B is used for requirements elicitation and task decomposition, while Qwendean 4B is employed; a modified version of Qwen3-4B fine-tuned for domain-specific UI code generation. A new iterative dataset creation procedure based on the combination of UI and TailwindCSS component ecosystems from ShadCN has been developed. Following several rounds of synthetic augmentation by means of DeepSeek-V3, the size of the created dataset increased from about 900 components to over 4,100 prompt-completion samples available under iamdyeus/ui-instruct-4k SLM fine-tuning was done using LoRA approach, supplemented by rsLoRA stability technique in 16-bit floating point arithmetic. The trained model showed its capability to synthesize cohesive interfaces based on React/TypeScript UI code and ShadCN/TailwindCSS components. A full-fledged heterogeneous pipeline based on LangGraph, consisting of orchestration, execution, and validation layers, was created. Experiments demonstrated an increase in the output quality consistency and lower inference time compared to the homogeneous GPT-LLM-based approach; in particular, about 3-5× lower latency was obtained within local section-based generation. Training artifacts, the agent harness, and an Electron application for local testing are available in the open-source repository iamDyeus/qwendean

Leer artículo completoexternamente

Me gusta

Guardar

Ver artículo completo