LVOmniBench: Pioneering Long Audio-Video Understanding Evaluation for Omnimodal LLMs

TL;DR

LVOmniBench introduces a comprehensive benchmark for evaluating long-form audio-visual understanding in omnimodal large language models, highlighting current models' limitations on extended videos and fostering future advancements.

Contribution

The paper presents LVOmniBench, a new dataset and evaluation framework for long-duration audio-visual comprehension in OmniLLMs, addressing a significant gap in existing short-clip benchmarks.

Findings

Current OmniLLMs struggle with long videos, achieving below 35% accuracy.

Gemini 3 Pro attains around 65% accuracy on the benchmark.

Long-form audio-visual understanding remains a challenging research area.

Abstract

Recent advancements in omnimodal large language models (OmniLLMs) have significantly improved the comprehension of audio and video inputs. However, current evaluations primarily focus on short audio and video clips ranging from 10 seconds to 5 minutes, failing to reflect the demands of real-world applications, where videos typically run for tens of minutes. To address this critical gap, we introduce LVOmniBench, a new benchmark designed specifically for the cross-modal comprehension of long-form audio and video. This dataset comprises high-quality videos sourced from open platforms that feature rich audio-visual dynamics. Through rigorous manual selection and annotation, LVOmniBench comprises 275 videos, ranging in duration from 10 to 90 minutes, and 1,014 question-answer (QA) pairs. LVOmniBench aims to rigorously evaluate the capabilities of OmniLLMs across domains, including long-term memory, temporal localization, fine-grained understanding, and multimodal perception. Our extensive evaluation reveals that current OmniLLMs encounter significant challenges when processing extended audio-visual inputs. Open-source models generally achieve accuracies below 35%, whereas the Gemini 3 Pro reaches a peak accuracy of approximately 65%. We anticipate that this dataset, along with our empirical findings, will stimulate further research and the development of advanced models capable of resolving complex cross-modal understanding problems within long-form audio-visual contexts.