# TAKE 85: Testing Audiovisual filmmaKer's intEnt across 85 Hours of Film

arXiv: https://arxiv.org/abs/2608.30068 · HTML: https://mlsanity.com/paper/2608.30068/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 44 of its 59 references that are in the map.
- First posted on arXiv 30 Aug 2026.

## Builds on (most cited first, 30 of 44)

- [GPT-4 Technical Report](https://mlsanity.com/paper/2303.08774/) (2023)
- [Qwen2-VL: Enhancing Vision-Language Model's Perception of the World at Any Resolution](https://mlsanity.com/paper/2409.12191/) (2024)
- [Robust Speech Recognition via Large-Scale Weak Supervision](https://mlsanity.com/paper/2212.04356/) (2022)
- [Qwen3-VL Technical Report](https://mlsanity.com/paper/2511.21631/) (2025)
- [InternVL3.5: Advancing Open-Source Multimodal Models in Versatility, Reasoning, and Efficiency](https://mlsanity.com/paper/2508.18265/) (2025)
- [Dense-Captioning Events in Videos](https://mlsanity.com/paper/1705.00754/) (2017)
- [ActivityNet-QA: A Dataset for Understanding Complex Web Videos via Question Answering](https://mlsanity.com/paper/1906.02467/) (2019)
- [VideoLLaMA 3: Frontier Multimodal Foundation Models for Image and Video Understanding](https://mlsanity.com/paper/2501.13106/) (2025)
- [TVQA: Localized, Compositional Video Question Answering](https://mlsanity.com/paper/1809.01696/) (2018)
- [MovieQA: Understanding Stories in Movies through Question-Answering](https://mlsanity.com/paper/1512.02902/) (2015)
- [Vid2Seq: Large-Scale Pretraining of a Visual Language Model for Dense Video Captioning](https://mlsanity.com/paper/2302.14115/) (2023)
- [Gemma 4 Technical Report](https://mlsanity.com/paper/2607.02770/) (2026)
- [Audio Flamingo 3: Advancing Audio Intelligence with Fully Open Large Audio Language Models](https://mlsanity.com/paper/2507.08128/) (2025)
- [End-to-End Dense Video Captioning with Masked Transformer](https://mlsanity.com/paper/1804.00819/) (2018)
- [MovieNet: A Holistic Dataset for Movie Understanding](https://mlsanity.com/paper/2007.10937/) (2020)
- [Video Understanding with Large Language Models: A Survey](https://mlsanity.com/paper/2312.17432/) (2023)
- [Towards Long-Form Video Understanding](https://mlsanity.com/paper/2106.11310/) (2021)
- [CinePile: A Long Video Question Answering Dataset and Benchmark](https://mlsanity.com/paper/2405.08813/) (2024)
- [Video ReCap: Recursive Captioning of Hour-Long Videos](https://mlsanity.com/paper/2402.13250/) (2024)
- [Streaming Dense Video Captioning](https://mlsanity.com/paper/2404.01297/) (2024)
- [Bidirectional Attentive Fusion with Context Gating for Dense Video Captioning](https://mlsanity.com/paper/1804.00100/) (2018)
- [Towards Understanding Camera Motions in Any Video](https://mlsanity.com/paper/2504.15376/) (2025)
- [From Representation to Reasoning: Towards both Evidence and Commonsense Reasoning for Video Question-Answering](https://mlsanity.com/paper/2205.14895/) (2022)
- [AutoAD: Movie Description in Context](https://mlsanity.com/paper/2303.16899/) (2023)
- [AutoAD II: The Sequel -- Who, When, and What in Movie Audio Description](https://mlsanity.com/paper/2310.06838/) (2023)
- [MM-Narrator: Narrating Long-form Videos with Multimodal In-Context Learning](https://mlsanity.com/paper/2311.17435/) (2023)
- [A Unified Framework for Shot Type Classification Based on Subject Centric Lens](https://mlsanity.com/paper/2008.03548/) (2020)
- [AutoAD III: The Prequel -- Back to the Pixels](https://mlsanity.com/paper/2404.14412/) (2024)
- [SAM Audio: Segment Anything in Audio](https://mlsanity.com/paper/2512.18099/) (2025)
- [AutoAD-Zero: A Training-Free Framework for Zero-Shot Audio Description](https://mlsanity.com/paper/2407.15850/) (2024)

## Abstract (arXiv)

Films communicate through deliberate creative choices, including lighting, color, composition, editing, dialogue, music, and sound. Humans naturally interpret these signals as directorial intent, yet current multimodal large language models (MLLMs) are evaluated almost exclusively on understanding what happens rather than why it is presented that way. We introduce TAKE 85, the first benchmark for directorial-intent understanding, comprising 398 short films (85 hours) with expert-verified question-answer pairs spanning global and fine-grained visual and audio intent. Through controlled modality ablations, TAKE 85 enables systematic evaluation of multimodal reasoning. Experiments on state-of-the-art MLLMs reveal a substantial gap between perceptual recognition and intentional understanding: while models accurately describe events and narratives, they consistently fail to infer the communicative role of filmmaking decisions. Our results establish directorial intent as a previously overlooked dimension of multimodal understanding: even the strongest model reaches only 58 out of 100, and our ablations show that no input modality is sufficient on its own. All code, Q&As, and models are publicly available from https://github.com/KaiShinozakiConefrey/Take-85