ailab-kyunghee

[IPCAI 2026] Official code for "SurgCheck: Do Vision–Language Models Really Look at Images in Surgical VQA?"

[ECCV 2026] ReQuest: Rethinking-based Question-Aware Frame Selection for Long-Form Video QA

[MICCAI 2025] Official code for "Towards Holistic Surgical Scene Graph"

[MICCAI 2025] Official code for "SurgX: Neuron–Concept Association for Explainable Surgical Phase Recognition"

[AAAI 2025]HiCM²: Hierarchical Compact Memory Modeling for Dense Video Captioning

Official implementation of "LLaVA Needs More Knowledge: Retrieval-Augmented NLG with Knowledge Graph for Explaining Thoracic Pathologies" (AAAI 2025). This repo provides the KG-LLaVA framework, integrating KG-based Retrieval-Augmented Generation (KG-RAG) with Vision-Language Models to generate NLEs for medical imaging.

[CVPR 2024] Do you remember? Dense Video Captioning with Cross-Modal Memory Retrieval