AdditiveLLM2: A Multi-modal Large Language Model for Additive Manufacturing

2026-03-23

AI Summary

This work addresses the limited domain expertise and multimodal comprehension of large language models in additive manufacturing by proposing a low-cost, efficient domain adaptation strategy. Building upon the Gemma-3 instruction-tuning architecture, the approach leverages a domain-specific dataset comprising approximately 50 million tokens from open-access additive manufacturing publications, integrating domain-adaptive pretraining, vision-instruction tuning, and multimodal fusion techniques to develop the first multimodal large language model tailored for additive manufacturing. The study also introduces Additive-Manufacturing-Benchmark, the first dedicated evaluation benchmark for this domain. Experimental results demonstrate that the proposed model achieves over 90% accuracy on general additive manufacturing knowledge tasks and exhibits strong performance across both linguistic and visual modalities.

Technology Category

• Machine Learning: Large Multimodal Models (LMMs)
• Cognitive Modeling & Cognitive Systems: Computational Creativity
• Natural Language Processing: Language Grounding & Multi-modal NLP

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Search and Retrieval-Augmented AI: Vertical and domain-specific search
• User Modeling, Personalization and Recommendation: Large Language Models (LLM) for user modeling and recommendation

Abstract

This work presents AdditiveLLM2 a multi-modal, domain adapted large language model built upon the instruction tuned variant of the Gemma 3 model using a relatively small dataset of around 50 million tokens. The dataset (AdditiveLLM2-OA) consists of open-access additive manufacturing journal articles with data extracted for the domain adaptive pretraining and visual instruction tuning processes. Various stages of the developed model are evaluated with the Additive-Manufacturing-Benchmark which consists of additive manufacturing domain specific tasks compiled published resources. AdditiveLLM2 exhibits proficiency in both language and vision based tasks, achieving accuracies upwards of 90% in general additive manufacturing knowledge. This domain adaptive pretraining and instruction tuning strategy outline an accessible specialization method for large language models to a domain such as additive manufacturing.

Problem

• Additive Manufacturing
• Large Language Model
• Multi-modal
• Domain Adaptation
• Instruction Tuning

Innovation

• multi-modal large language model
• domain adaptive pretraining
• instruction tuning
• additive manufacturing
• vision-language tasks