FALCON: Functional Assembly and Language for Compositional Reasoning in X-ray

Conventional vision-language models are largely object-centric, focusing on detecting and describing individual entities. In safety-critical X-ray baggage screening, however, risk often emerges not from a single object but from the functional compatibility of spatially dispersed, dismantled IED components such as batteries, detonators, and explosive charges. We formalize this setting as compositional threat reasoning, where risk is modeled as a relational property of grounded regions rather than an independent detection outcome. We introduce FALCON, a multimodal framework for reasoning and grounding over dismantled IED components. FALCON abstracts segmentation-aware region features into a structured safety state capturing component presence, pairwise functional compatibility, and scene-level risk. This structured representation is injected into the language model as an explicit intermediate interface, encouraging relationally consistent and safety-aware reasoning. Falcon-X evaluates dense grounding with structured supervision over dismantled component completeness, compatibility, and risk inference in cluttered X-ray imagery. Experiments show that existing multimodal models adapt to appearance but struggle with compositional safety reasoning, while FALCON improves functional grounding and produces more coherent scene-level risk assessments.