LVSum: A Benchmark for Timestamp-Aware Long Video Summarization

TL;DR

LVSum is a new benchmark dataset with human annotations for evaluating long video summarization, emphasizing temporal fidelity and multimodal grounding across diverse domains.

Contribution

The paper introduces LVSum, a benchmark with fine-grained temporal annotations, and evaluates current models, highlighting gaps in temporal understanding for long videos.

Findings

Existing MLLMs show systematic gaps in temporal understanding.

LVSum enables detailed evaluation of temporal and semantic alignment.

New metrics reveal limitations of current models in long video summarization.

Abstract

Long video summarization presents significant challenges for current multimodal large language models (MLLMs), particularly in maintaining temporal fidelity over extended durations and producing summaries that are both semantically and temporally grounded. In this work, we present LVSum, a human-annotated benchmark designed specifically for evaluating long video summarization with fine-grained temporal alignment. LVSum comprises diverse long-form videos across 13 domains, each paired with human-generated summaries containing precise temporal references. We conduct a comprehensive evaluation of both proprietary and open-source MLLMs on LVSum, assessing performance using newly introduced LLM-based metrics for content relevance and modality coherence, alongside standard evaluation metrics. Our experiments reveal systematic gaps in temporal understanding among existing MLLMs and offer insights that establish a new foundation for advancing temporal reasoning in long video summarization.