# FIOVA: A Multi-Annotator Benchmark for Human-Aligned Video Captioning

arXiv: https://arxiv.org/abs/2410.15270 · HTML: https://mlsanity.com/paper/2410.15270/

## Key facts

- Cited by 7 papers in the map (data through 14 Oct 2026).
- Placed in Video understanding with MLLMs, one of 16 directions in Multimodal LLMs / vision-language models.
- First posted on arXiv 20 Oct 2024.

## Placed in

- Multimodal LLMs / vision-language models / [Video understanding with MLLMs](https://mlsanity.com/mllm/video-understanding-with-mllms/)

## Built on by (most cited first, 7 of 7)

- [CausalStep: A Benchmark for Explicit Stepwise Causal Reasoning in Videos](https://mlsanity.com/paper/2507.16878/) (2025)
- [Look Less, Reason More: Rollout-Guided Adaptive Pixel-Space Reasoning](https://mlsanity.com/paper/2510.01681/) (2025)
- [Select Less, Reason More: Prioritizing Evidence Purity for Video Reasoning](https://mlsanity.com/paper/2510.15440/) (2025)
- [Beyond Accuracy: Evaluating Grounded Visual Evidence in Thinking with Images](https://mlsanity.com/paper/2601.11633/) (2026)
- [CapRiCorn-1K: A Comprehensive Benchmark for Video Captioning and Subject Referential Consistency Across Temporal Scales](https://mlsanity.com/paper/2606.21949/) (2026)
- [MathMixup: Boosting LLM Mathematical Reasoning with Difficulty-Controllable Data Synthesis and Curriculum Learning](https://mlsanity.com/paper/2601.17006/) (2026)
- [A Survey of Large Models in Sports](https://mlsanity.com/paper/2608.14377/) (2026)

## Abstract (arXiv)

Despite rapid progress in large vision-language models (LVLMs), existing video caption benchmarks remain limited in evaluating their alignment with human understanding. Most rely on a single annotation per video and lexical similarity-based metrics, failing to capture the variability in human perception and the cognitive importance of events. These limitations hinder accurate diagnosis of model capabilities in producing coherent, complete, and human-aligned descriptions. To address this, we introduce FIOVA (Five-In-One Video Annotations), a human-centric benchmark tailored for evaluation. It comprises 3,002 real-world videos (about 33.6s each), each annotated independently by five annotators. This design enables modeling of semantic diversity and inter-subjective agreement, offering a richer foundation for measuring human-machine alignment. We further propose FIOVA-DQ, an event-level evaluation metric that incorporates cognitive weights derived from annotator consensus, providing fine-grained assessment of event relevance and semantic coverage. Leveraging FIOVA, we conduct a comprehensive evaluation of nine representative LVLMs and introduce a complexity-aware analysis framework based on inter-annotator variation (CV). This reveals consistency gaps across difficulty levels and identifies structural issues such as event under-description and template convergence. Our results highlight FIOVA's diagnostic value for understanding LVLM behavior under varying complexity, setting a new standard for cognitively aligned evaluation in long-video captioning. The benchmark, annotations, metric, and model outputs are publicly released to support future evaluation-driven research in video understanding. More detailed information can be found at https://huuuuusy.github.io/fiova/.