Search papers, labs, and topics across Lattice.
The paper introduces FORGE, a new multimodal dataset combining real-world 2D images and 3D point clouds with fine-grained domain semantics for manufacturing scenarios, designed to address the limitations of current MLLM evaluations in reflecting real-world manufacturing demands. Evaluating 18 state-of-the-art MLLMs on workpiece verification, structural surface inspection, and assembly verification tasks revealed that insufficient domain-specific knowledge, rather than visual grounding, is the primary bottleneck. Supervised fine-tuning of a 3B-parameter model on the FORGE dataset achieved up to 90.8% relative accuracy improvement, demonstrating a practical pathway toward domain-adapted manufacturing MLLMs.
Turns out, MLLMs struggle with manufacturing tasks not because they can't "see," but because they lack the domain-specific knowledge to understand what they're looking at.
The manufacturing sector is increasingly adopting Multimodal Large Language Models (MLLMs) to transition from simple perception to autonomous execution, yet current evaluations fail to reflect the rigorous demands of real-world manufacturing environments. Progress is hindered by data scarcity and a lack of fine-grained domain semantics in existing datasets. To bridge this gap, we introduce FORGE. Wefirst construct a high-quality multimodal dataset that combines real-world 2D images and 3D point clouds, annotated with fine-grained domain semantics (e.g., exact model numbers). We then evaluate 18 state-of-the-art MLLMs across three manufacturing tasks, namely workpiece verification, structural surface inspection, and assembly verification, revealing significant performance gaps. Counter to conventional understanding, the bottleneck analysis shows that visual grounding is not the primary limiting factor. Instead, insufficient domain-specific knowledge is the key bottleneck, setting a clear direction for future research. Beyond evaluation, we show that our structured annotations can serve as an actionable training resource: supervised fine-tuning of a compact 3B-parameter model on our data yields up to 90.8% relative improvement in accuracy on held-out manufacturing scenarios, providing preliminary evidence for a practical pathway toward domain-adapted manufacturing MLLMs. The code and datasets are available at https://ai4manufacturing.github.io/forge-web.
