MINED: Probing and Updating with Multimodal Time-Sensitive Knowledge for Large Multimodal Models

TL;DR

This paper introduces MINED, a benchmark for evaluating large multimodal models' understanding of time-sensitive knowledge, revealing current limitations and exploring knowledge updating methods.

Contribution

The paper presents MINED, a new benchmark with 2,104 samples to evaluate temporal awareness in LMMs and investigates knowledge editing for updates.

Findings

Gemini-2.5-Pro achieves the highest CEM score of 63.07.

Most open-source LMMs lack strong time-sensitive knowledge understanding.

LMMs can effectively update knowledge via editing in single scenarios.

Abstract

Large Multimodal Models (LMMs) encode rich factual knowledge via cross-modal pre-training, yet their static representations struggle to maintain an accurate understanding of time-sensitive factual knowledge. Existing benchmarks remain constrained by static designs, inadequately evaluating LMMs' ability to understand time-sensitive knowledge. To address this gap, we propose MINED, a comprehensive benchmark that evaluates temporal awareness along 6 key dimensions and 11 challenging tasks: cognition, awareness, trustworthiness, understanding, reasoning, and robustness. MINED is constructed from Wikipedia by two professional annotators, containing 2,104 time-sensitive knowledge samples spanning six knowledge types. Evaluating 15 widely used LMMs on MINED shows that Gemini-2.5-Pro achieves the highest average CEM score of 63.07, while most open-source LMMs still lack time understanding ability. Meanwhile, LMMs perform best on organization knowledge, whereas their performance is weakest on sport. To address these challenges, we investigate the feasibility of updating time-sensitive knowledge in LMMs through knowledge editing methods and observe that LMMs can effectively update knowledge via knowledge editing methods in single editing scenarios.