# ShotBench: Expert-Level Cinematic Understanding in Vision-Language Models

arXiv: https://arxiv.org/abs/2506.21356 · HTML: https://mlsanity.com/paper/2506.21356/

## Key facts

- Cited by 22 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 41 of its 59 references that are in the map.
- First posted on arXiv 26 Jun 2025.

## 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 41)

- [DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models](https://mlsanity.com/paper/2402.03300/) (2024)
- [Qwen2.5-VL Technical Report](https://mlsanity.com/paper/2502.13923/) (2025)
- [LLaVA-OneVision: Easy Visual Task Transfer](https://mlsanity.com/paper/2408.03326/) (2024)
- [FlashAttention-2: Faster Attention with Better Parallelism and Work Partitioning](https://mlsanity.com/paper/2307.08691/) (2023)
- [CogVideoX: Text-to-Video Diffusion Models with An Expert Transformer](https://mlsanity.com/paper/2408.06072/) (2024)
- [MMBench: Is Your Multi-modal Model an All-around Player?](https://mlsanity.com/paper/2307.06281/) (2023)
- [InternVL3: Exploring Advanced Training and Test-Time Recipes for Open-Source Multimodal Models](https://mlsanity.com/paper/2504.10479/) (2025)
- [HunyuanVideo: A Systematic Framework For Large Video Generative Models](https://mlsanity.com/paper/2412.03603/) (2024)
- [Expanding Performance Boundaries of Open-Source Multimodal Models with Model, Data, and Test-Time Scaling](https://mlsanity.com/paper/2412.05271/) (2024)
- [Eyes Wide Shut? Exploring the Visual Shortcomings of Multimodal LLMs](https://mlsanity.com/paper/2401.06209/) (2024)
- [DataComp: In search of the next generation of multimodal datasets](https://mlsanity.com/paper/2304.14108/) (2023)
- [Thinking in Space: How Multimodal Large Language Models See, Remember, and Recall Spaces](https://mlsanity.com/paper/2412.14171/) (2024)
- [Visual-RFT: Visual Reinforcement Fine-Tuning](https://mlsanity.com/paper/2503.01785/) (2025)
- [VLMEvalKit: An Open-Source Toolkit for Evaluating Large Multi-Modality Models](https://mlsanity.com/paper/2407.11691/) (2024)
- [InternLM-XComposer2: Mastering Free-form Text-Image Composition and Comprehension in Vision-Language Large Model](https://mlsanity.com/paper/2401.16420/) (2024)
- [VL-Rethinker: Incentivizing Self-Reflection of Vision-Language Models with Reinforcement Learning](https://mlsanity.com/paper/2504.08837/) (2025)
- [ReCamMaster: Camera-Controlled Generative Rendering from A Single Video](https://mlsanity.com/paper/2503.11647/) (2025)
- [LogicVista: Multimodal LLM Logical Reasoning Benchmark in Visual Contexts](https://mlsanity.com/paper/2407.04973/) (2024)
- [Video-MMMU: Evaluating Knowledge Acquisition from Multi-Discipline Professional Videos](https://mlsanity.com/paper/2501.13826/) (2025)
- [CharXiv: Charting Gaps in Realistic Chart Understanding in Multimodal LLMs](https://mlsanity.com/paper/2406.18521/) (2024)
- [DigiRL: Training In-The-Wild Device-Control Agents with Autonomous Reinforcement Learning](https://mlsanity.com/paper/2406.11896/) (2024)
- [MovieNet: A Holistic Dataset for Movie Understanding](https://mlsanity.com/paper/2007.10937/) (2020)
- [TransNet V2: An effective deep network architecture for fast shot transition detection](https://mlsanity.com/paper/2008.04838/) (2020)
- [Ovis: Structural Embedding Alignment for Multimodal Large Language Model](https://mlsanity.com/paper/2405.20797/) (2024)
- [PhysBench: Benchmarking and Enhancing Vision-Language Models for Physical World Understanding](https://mlsanity.com/paper/2501.16411/) (2025)
- [Oryx MLLM: On-Demand Spatial-Temporal Understanding at Arbitrary Resolution](https://mlsanity.com/paper/2409.12961/) (2024)
- [MMVU: Measuring Expert-Level Multi-Discipline Video Understanding](https://mlsanity.com/paper/2501.12380/) (2025)
- [Insight-V: Exploring Long-Chain Visual Reasoning with Multimodal Large Language Models](https://mlsanity.com/paper/2411.14432/) (2024)
- [EgoLife: Towards Egocentric Life Assistant](https://mlsanity.com/paper/2503.03803/) (2025)
- [Ola: Pushing the Frontiers of Omni-Modal Language Model](https://mlsanity.com/paper/2502.04328/) (2025)

## Built on by (most cited first, 22 of 22)

- [OneStory: Coherent Multi-Shot Video Generation with Adaptive Memory](https://mlsanity.com/paper/2512.07802/) (2025)
- [Thinking with Camera: A Unified Multimodal Model for Camera-Centric Understanding and Generation](https://mlsanity.com/paper/2510.08673/) (2025)
- [Cut2Next: Generating Next Shot via In-Context Tuning](https://mlsanity.com/paper/2508.08244/) (2025)
- [RefineShot: Rethinking Cinematography Understanding with Foundational Skill Evaluation](https://mlsanity.com/paper/2510.02423/) (2025)
- [CamReasoner: Reinforcing Camera Movement Understanding via Structured Spatial Reasoning](https://mlsanity.com/paper/2602.00181/) (2026)
- [AgenticVBench: Can AI Agents Complete Real-World Post-Production Tasks?](https://mlsanity.com/paper/2605.27705/) (2026)
- [OmniShotCut: Holistic Relational Shot Boundary Detection with Shot-Query Transformer](https://mlsanity.com/paper/2604.24762/) (2026)
- [GS-Agent: Creating 4D Physical Worlds With Generative Simulation](https://mlsanity.com/paper/2607.21522/) (2026)
- [Seeing without Pixels: Perception from Camera Trajectories](https://mlsanity.com/paper/2511.21681/) (2025)
- [Camera Artist: A Multi-Agent Framework for Cinematic Language Storytelling Video Generation](https://mlsanity.com/paper/2604.09195/) (2026)
- [Geometry-Guided Camera Motion Understanding in VideoLLMs](https://mlsanity.com/paper/2603.13119/) (2026)
- [VERTIGO: Visual Preference Optimization for Cinematic Camera Trajectory Generation](https://mlsanity.com/paper/2604.02467/) (2026)
- [VEBench:Benchmarking Large Multimodal Models for Real-World Video Editing](https://mlsanity.com/paper/2605.03276/) (2026)
- [CineCap: Structured Reasoning with Spatio-Temporal Anchors for Cinematographic Video Captioning](https://mlsanity.com/paper/2606.24636/) (2026)
- [CutVerse: A Compositional GUI Agents Benchmark for Media Post-Production Editing](https://mlsanity.com/paper/2605.19484/) (2026)
- [Natural Language Camera Movement Understanding](https://mlsanity.com/paper/2607.03043/) (2026)
- [Temporally Grounded Compositional Camera Motion Understanding via Geometric Knowledge Distillation](https://mlsanity.com/paper/2608.10932/) (2026)
- [Controlling Embedding Spaces with Text-Conditioned Transformations](https://mlsanity.com/paper/2607.22919/) (2026)
- [TAKE 85: Testing Audiovisual filmmaKer's intEnt across 85 Hours of Film](https://mlsanity.com/paper/2608.30068/) (2026)
- [DramaChain Bench: An End-to-End Benchmark for Short-Drama Generation](https://mlsanity.com/paper/2609.00646/) (2026)
- [CinematicVQA: Benchmarking Film-Grammar Reasoning in Large Vision-Language Models](https://mlsanity.com/paper/2609.28813/) (2026)
- [Unveiling the Value of Motion for Cinematic Camera Trajectories](https://mlsanity.com/paper/2609.38683/) (2026)

## Abstract (arXiv)

Cinematography, the fundamental visual language of film, is essential for conveying narrative, emotion, and aesthetic quality. While recent Vision-Language Models (VLMs) demonstrate strong general visual understanding, their proficiency in comprehending the nuanced cinematic grammar embedded within individual shots remains largely unexplored and lacks robust evaluation. This critical gap limits both fine-grained visual comprehension and the precision of AI-assisted video generation. To address this, we introduce ShotBench, a comprehensive benchmark specifically designed for cinematic language understanding. It features over 3.5k expert-annotated QA pairs from images and video clips, meticulously curated from over 200 acclaimed (predominantly Oscar-nominated) films and spanning eight key cinematography dimensions. Our evaluation of 24 leading VLMs on ShotBench reveals their substantial limitations: even the top-performing model achieves less than 60% average accuracy, particularly struggling with fine-grained visual cues and complex spatial reasoning. To catalyze advancement in this domain, we construct ShotQA, a large-scale multimodal dataset comprising approximately 70k cinematic QA pairs. Leveraging ShotQA, we develop ShotVL through supervised fine-tuning and Group Relative Policy Optimization. ShotVL significantly outperforms all existing open-source and proprietary models on ShotBench, establishing new state-of-the-art performance. We open-source our models, data, and code to foster rapid progress in this crucial area of AI-driven cinematic understanding and generation.