{"ID":2867976,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2509.18436","arxiv_id":"2509.18436","title":"Memory-QA: Answering Recall Questions Based on Multimodal Memories","abstract":"We introduce Memory-QA, a novel real-world task that involves answering recall questions about visual content from previously stored multimodal memories. This task poses unique challenges, including the creation of task-oriented memories, the effective utilization of temporal and location information within memories, and the ability to draw upon multiple memories to answer a recall question. To address these challenges, we propose a comprehensive pipeline, Pensieve, integrating memory-specific augmentation, time- and location-aware multi-signal retrieval, and multi-memory QA fine-tuning. We created a multimodal benchmark to illustrate various real challenges in this task, and show the superior performance of Pensieve over state-of-the-art solutions (up to 14% on QA accuracy).","short_abstract":"We introduce Memory-QA, a novel real-world task that involves answering recall questions about visual content from previously stored multimodal memories. This task poses unique challenges, including the creation of task-oriented memories, the effective utilization of temporal and location information within memories, a...","url_abs":"https://arxiv.org/abs/2509.18436","url_pdf":"https://arxiv.org/pdf/2509.18436v2","authors":"[\"Hongda Jiang\",\"Xinyuan Zhang\",\"Siddhant Garg\",\"Rishab Arora\",\"Shiun-Zu Kuo\",\"Jiayang Xu\",\"Ankur Bansal\",\"Christopher Brossman\",\"Yue Liu\",\"Aaron Colak\",\"Ahmed Aly\",\"Anuj Kumar\",\"Xin Luna Dong\"]","published":"2025-09-22T21:41:35Z","proceeding":"cs.AI","tasks":"[\"cs.AI\",\"cs.CL\",\"cs.DB\"]","methods":"[]","has_code":false}