2025/06/19 by Lando, Giuseppe, Forte, Rosario, Farinella, Giovanni Maria +1
Computer Science · #Computer Vision and Pattern Recognition (cs.CV) #Domain Adaptation and Few-Shot Learning #FOS: Computer and information sciences #Multimodal Machine Learning Applications #Topic Modeling
paper · pdf · doi:10.48550/arxiv.2506.16450
openalex publication_date 2025/06/19 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28
We investigate whether off-the-shelf Multimodal Large Language Models (MLLMs) can tackle Online Episodic-Memory Video Question Answering (OEM-VQA) without additional training. Our pipeline converts a streaming egocentric video into a lightweight textual memory, only a few kilobytes per minute, via an MLLM descriptor module, and answers multiple-choice questions by querying this memory with an LLM reasoner module. On the QAEgo4D-Closed benchmark, our best configuration attains 56.0% accuracy with 3.6 kB per minute storage, matching the performance of dedicated state-of-the-art systems while being 10**4/10**5 times more memory-efficient. Extensive ablations provides insights into the role of each component and design choice, and highlight directions of improvement for future research.