# Script-a-Video: Deep Structured Audio-visual Captions via Factorized Streams and Relational Grounding

arXiv: https://arxiv.org/abs/2604.11244 · HTML: https://mlsanity.com/paper/2604.11244/

## Key facts

- Cited by 8 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.
- Builds on 67 of its 74 references that are in the map.
- First posted on arXiv 13 Apr 2026.

## Placed in

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

## Builds on (most cited first, 30 of 67)

- [DINOv2: Learning Robust Visual Features without Supervision](https://mlsanity.com/paper/2304.07193/) (2023)
- [Robust Speech Recognition via Large-Scale Weak Supervision](https://mlsanity.com/paper/2212.04356/) (2022)
- [Video-LLaMA: An Instruction-tuned Audio-Visual Language Model for Video Understanding](https://mlsanity.com/paper/2306.02858/) (2023)
- [Video-ChatGPT: Towards Detailed Video Understanding via Large Vision and Language Models](https://mlsanity.com/paper/2306.05424/) (2023)
- [VideoChat: Chat-Centric Video Understanding](https://mlsanity.com/paper/2305.06355/) (2023)
- [Qwen3-Omni Technical Report](https://mlsanity.com/paper/2509.17765/) (2025)
- [UTMOS: UTokyo-SaruLab System for VoiceMOS Challenge 2022](https://mlsanity.com/paper/2204.02152/) (2022)
- [StoryDiffusion: Consistent Self-Attention for Long-Range Image and Video Generation](https://mlsanity.com/paper/2405.01434/) (2024)
- [LTX-2: Efficient Joint Audio-Visual Foundation Model](https://mlsanity.com/paper/2601.03233/) (2026)
- [DreamVideo: Composing Your Dream Videos with Customized Subject and Motion](https://mlsanity.com/paper/2312.04433/) (2023)
- [Identity-Preserving Text-to-Video Generation by Frequency Decomposition](https://mlsanity.com/paper/2411.17440/) (2024)
- [WorldSense: Evaluating Real-world Omnimodal Understanding for Multimodal LLMs](https://mlsanity.com/paper/2502.04326/) (2025)
- [Phantom: Subject-consistent video generation via cross-modal alignment](https://mlsanity.com/paper/2502.11079/) (2025)
- [Long Context Tuning for Video Generation](https://mlsanity.com/paper/2503.10589/) (2025)
- [Baichuan-Omni-1.5 Technical Report](https://mlsanity.com/paper/2501.15368/) (2025)
- [Daily-Omni: Towards Audio-Visual Reasoning with Temporal Alignment across Modalities](https://mlsanity.com/paper/2505.17862/) (2025)
- [video-SALMONN: Speech-Enhanced Audio-Visual Large Language Models](https://mlsanity.com/paper/2406.15704/) (2024)
- [R1-Omni: Explainable Omni-Multimodal Emotion Recognition with Reinforcement Learning](https://mlsanity.com/paper/2503.05379/) (2025)
- [Mixture of Contexts for Long Video Generation](https://mlsanity.com/paper/2508.21058/) (2025)
- [HunyuanCustom: A Multimodal-Driven Architecture for Customized Video Generation](https://mlsanity.com/paper/2505.04512/) (2025)
- [Ola: Pushing the Frontiers of Omni-Modal Language Model](https://mlsanity.com/paper/2502.04328/) (2025)
- [Ovi: Twin Backbone Cross-Modal Fusion for Audio-Video Generation](https://mlsanity.com/paper/2510.01284/) (2025)
- [CustomVideo: Customizing Text-to-Video Generation with Multiple Subjects](https://mlsanity.com/paper/2401.09962/) (2024)
- [SkyReels-A2: Compose Anything in Video Diffusion Transformers](https://mlsanity.com/paper/2504.02436/) (2025)
- [ConceptMaster: Multi-Concept Video Customization on Diffusion Transformer Models Without Test-Time Tuning](https://mlsanity.com/paper/2501.04698/) (2025)
- [Multi-subject Open-set Personalization in Video Generation](https://mlsanity.com/paper/2501.06187/) (2025)
- [UniVerse-1: Unified Audio-Video Generation via Stitching of Experts](https://mlsanity.com/paper/2509.06155/) (2025)
- [OmniVideoBench: Towards Audio-Visual Understanding Evaluation for Omni MLLMs](https://mlsanity.com/paper/2510.10689/) (2025)
- [HumanOmniV2: From Understanding to Omni-Modal Reasoning with Context](https://mlsanity.com/paper/2506.21277/) (2025)
- [HoloCine: Holistic Generation of Cinematic Multi-Shot Long Video Narratives](https://mlsanity.com/paper/2510.20822/) (2025)

## Built on by (most cited first, 8 of 8)

- [OmniVideo-100K: A Dataset for Audio-Visual Reasoning through Structured Scripts and Evidence Chains](https://mlsanity.com/paper/2606.14702/) (2026)
- [CapRiCorn-1K: A Comprehensive Benchmark for Video Captioning and Subject Referential Consistency Across Temporal Scales](https://mlsanity.com/paper/2606.21949/) (2026)
- [Aura: Consistent Multi-Subject Video Generation via VLM-Grounded Semantic Alignment](https://mlsanity.com/paper/2607.04311/) (2026)
- [SARA: Semantically Adaptive Relational Alignment for Video Diffusion Models](https://mlsanity.com/paper/2605.07800/) (2026)
- [The Missing Temporal Link: Temporal Context Routing for Script-Driven Audio-Video Generation](https://mlsanity.com/paper/2609.02367/) (2026)
- [OmniCapBench: A Deep-Structured Evaluation Framework for Fine-Grained Audio-Visual Captioning](https://mlsanity.com/paper/2610.12458/) (2026)
- [Structured Stateful Relational Scripts with a High-Quality Dataset for Multi-Shot Audio-Visual Video Generation](https://mlsanity.com/paper/2610.15512/) (2026)
- [PIGEONS: Incremental Long-Video Captioning with Persistent Identity Gallery and Evolving Overarching Narrative Summary](https://mlsanity.com/paper/2610.16066/) (2026)

## Abstract (arXiv)

Advances in Multimodal Large Language Models (MLLMs) are transforming video captioning from a descriptive endpoint into a semantic interface for both video understanding and generation. However, the dominant paradigm still casts videos as monolithic narrative paragraphs that entangle visual, auditory, and identity information. This dense coupling not only compromises representational fidelity but also limits scalability, since even local edits can trigger global rewrites. To address this structural bottleneck, we propose Multi-Stream Scene Script (MTSS), a novel paradigm that replaces monolithic text with factorized and explicitly grounded scene descriptions. MTSS is built on two core principles: Stream Factorization, which decouples a video into complementary streams (Reference, Shot, Event, and Global), and Relational Grounding, which reconnects these isolated streams through explicit identity and temporal links to maintain holistic video consistency. Extensive experiments demonstrate that MTSS consistently enhances video understanding across various models, achieving an average reduction of 25% in the total error rate on Video-SALMONN-2 and an average performance gain of 67% on the Daily-Omni reasoning benchmark. It also narrows the performance gap between smaller and larger MLLMs, indicating a substantially more learnable caption interface. Finally, even without architectural adaptation, replacing monolithic prompts with MTSS in multi-shot video generation yields substantial human-rated improvements: a 45% boost in cross-shot identity consistency, a 56% boost in audio-visual alignment, and a 71% boost in temporal controllability.