UniHEAR: Unified Heterogeneous-Source Attentive Retrieval for Knowledge-Based Visual Question Answering
UniHEAR is a unified lightweight framework for heterogeneous-source entity retrieval and reranking that achieves state-of-the-art retrieval and VQA performance, improving Recall@1 and Recall@1 by 6.7 and 1.2 points over the strongest baselines while maintaining a lightweight reranking architecture.