EAMA : Entity-Aware Multimodal Alignment Based Approach for News Image Captioning
This addresses the challenge of balancing sufficiency and conciseness in entity-rich captions for news image captioning, representing an incremental improvement over existing methods.
The paper tackles the problem of generating informative captions rich in entities for news images by proposing EAMA, an entity-aware multimodal alignment approach that aligns a multimodal large language model with extra tasks to supplement entity-related information, achieving better results than all previous models on two mainstream datasets.
News image captioning requires model to generate an informative caption rich in entities, with the news image and the associated news article. Current MLLMs still bear limitations in handling entity information in news image captioning tasks. Besides, generating high-quality news image captions requires a trade-off between sufficiency and conciseness of textual input information. To explore the potential of MLLMs and address problems we discovered, we propose EAMA: an Entity-Aware Multimodal Alignment based approach for News Image Captioning. Our approach first aligns the MLLM with two extra alignment tasks: Entity-Aware Sentence Selection task and Entity Selection task, together with News Image Captioning task. The aligned MLLM will utilize the additional entity-related information extracted by itself to supplement the textual input while generating news image captions. Our approach achieves better results than all previous models on two mainstream news image captioning datasets.