# How Far Can Off-the-Shelf Multimodal Large Language Models Go in Online Episodic Memory Question Answering?

Source: [How Far Can Off-the-Shelf Multimodal Large Language Models Go in Online Episodic Memory Question Answering?](<https://www.sota2.com/research/paper/68fa79911e209131cea7e483>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2025-10-28T13:57:18.278Z

Authors: Giuseppe Lando, Rosario Forte, Giovanni Maria Farinella, Antonino Furnari

Published: 2025-06-19T16:35:49.000Z

## Abstract

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.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

- [Video Question Answering on QAego4D Closed](<https://www.sota2.com/research/sota/video-question-answering-on-qaego4d-closed.md>): Video Question Answering; QAego4D Closed; Accuracy: 56.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2506.16450>)

[PDF](<https://arxiv.org/pdf/2506.16450.pdf>)