Search papers, labs, and topics across Lattice.
This paper introduces ViSR-KGC, a novel approach for multimodal knowledge graph completion that leverages visual subgraph reasoning to enhance entity and relation inference. By integrating representation learning for global topology, vision-language models for local multimodal evidence, and commonsense knowledge, the framework effectively captures complex semantic correlations in knowledge graphs. The key finding demonstrates that ViSR-KGC significantly improves completion accuracy compared to traditional embedding-based and linearized LLM methods, showcasing the importance of visual context in KGC tasks.
Visual context can dramatically enhance knowledge graph completion, as shown by ViSR-KGC's superior accuracy over traditional methods.
Knowledge graph completion (KGC) aims to infer missing entities or relations from incomplete graph structures, and has evolved into multimodal knowledge graph completion (MMKGC), where entities are associated with multiple modalities such as text and images. Traditional representation learning approaches follow the embedding-based paradigm and may struggle when relation-specific evidence is limited. Meanwhile, LLM-based reasoning methods typically linearize graph structures into textual prompts, which obscures structural topology and neglects vital visual information. While vision-language models (VLMs) excel at multimodal reasoning, they cannot natively interpret structured graph topology, particularly when it comes to knowledge graphs where nodes and edges carry complex semantics. To bridge this gap, we propose ViSR-KGC, a visual subgraph reasoning approach for KGC. It integrates three complementary capabilities to capture semantic correlations: identifying global topology dependencies via representation learning, analyzing local multimodal evidence using VLMs, and providing necessary commonsense knowledge inherent in pre-trained models. Based on learned multimodal embeddings, our framework first extracts a compact and query-aware subgraph from the MMKG. Then, this subgraph is transformed into a visually interpretable image using a layout strategy selected through empirical comparison.Finally, the visualized subgraph, entity images, textual descriptions, and candidate answers are combined into a unified prompt, enabling the VLM to infer the missing entity.
