LVOmniBench: Pioneering Long Audio-Video Understanding Evaluation for Omnimodal LLMs
2026-03-19 • Computer Vision and Pattern Recognition
Computer Vision and Pattern Recognition
AI summaryⓘ
The authors created a new test called LVOmniBench to check how well large language models understand long videos with both sound and pictures. Unlike past tests that only used short clips, this one uses longer videos, from 10 to 90 minutes, with questions to see how well the models remember and understand the details. They found that most current models struggle with these long videos, with only one model doing somewhat better but still not perfect. The authors hope this new test will help improve future models for understanding long videos better.
OmniLLMslong-form videocross-modal comprehensionaudio-visual dynamicsbenchmark datasettemporal localizationfine-grained understandingmultimodal perceptionquestion-answeringlong-term memory
Authors
Keda Tao, Yuhua Zheng, Jia Xu, Wenjie Du, Kele Shao, Hesong Wang, Xueyi Chen, Xin Jin, Junhan Zhu, Bohan Yu, Weiqiang Wang, Jian Liu, Can Qin, Yulun Zhang, Ming-Hsuan Yang, Huan Wang
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.