# ScaleLong: A Multi-Timescale Benchmark for Long Video Understanding

arXiv: https://arxiv.org/abs/2505.23922 · HTML: https://mlsanity.com/paper/2505.23922/

## Key facts

- Cited by 8 papers in the map (data through 14 Oct 2026).
- Builds on 30 of its 34 references that are in the map.
- First posted on arXiv 29 May 2025.

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

- [Visual Instruction Tuning](https://mlsanity.com/paper/2304.08485/) (2023)
- [Qwen2.5-VL Technical Report](https://mlsanity.com/paper/2502.13923/) (2025)
- [Gemini 1.5: Unlocking multimodal understanding across millions of tokens of context](https://mlsanity.com/paper/2403.05530/) (2024)
- [InternVL: Scaling up Vision Foundation Models and Aligning for Generic Visual-Linguistic Tasks](https://mlsanity.com/paper/2312.14238/) (2023)
- [InternVL3: Exploring Advanced Training and Test-Time Recipes for Open-Source Multimodal Models](https://mlsanity.com/paper/2504.10479/) (2025)
- [Expanding Performance Boundaries of Open-Source Multimodal Models with Model, Data, and Test-Time Scaling](https://mlsanity.com/paper/2412.05271/) (2024)
- [Video-LLaMA: An Instruction-tuned Audio-Visual Language Model for Video Understanding](https://mlsanity.com/paper/2306.02858/) (2023)
- [Video-MME: The First-Ever Comprehensive Evaluation Benchmark of Multi-modal LLMs in Video Analysis](https://mlsanity.com/paper/2405.21075/) (2024)
- [MVBench: A Comprehensive Multi-modal Video Understanding Benchmark](https://mlsanity.com/paper/2311.17005/) (2023)
- [mPLUG-Owl: Modularization Empowers Large Language Models with Multimodality](https://mlsanity.com/paper/2304.14178/) (2023)
- [NExT-QA:Next Phase of Question-Answering to Explaining Temporal Actions](https://mlsanity.com/paper/2105.08276/) (2021)
- [EgoSchema: A Diagnostic Benchmark for Very Long-form Video Language Understanding](https://mlsanity.com/paper/2308.09126/) (2023)
- [LongVideoBench: A Benchmark for Long-context Interleaved Video-Language Understanding](https://mlsanity.com/paper/2407.15754/) (2024)
- [Otter: A Multi-Modal Model with In-Context Instruction Tuning](https://mlsanity.com/paper/2305.03726/) (2023)
- [LLaMA-VID: An Image is Worth 2 Tokens in Large Language Models](https://mlsanity.com/paper/2311.17043/) (2023)
- [LVBench: An Extreme Long Video Understanding Benchmark](https://mlsanity.com/paper/2406.08035/) (2024)
- [Kimi-VL Technical Report](https://mlsanity.com/paper/2504.07491/) (2025)
- [mPLUG-Owl3: Towards Long Image-Sequence Understanding in Multi-Modal Large Language Models](https://mlsanity.com/paper/2408.04840/) (2024)
- [STAR: A Benchmark for Situated Reasoning in Real-World Videos](https://mlsanity.com/paper/2405.09711/) (2024)
- [Flash-VStream: Memory-Based Real-Time Understanding for Long Video Streams](https://mlsanity.com/paper/2406.08085/) (2024)
- [MAmmoTH-VL: Eliciting Multimodal Reasoning with Instruction Tuning at Scale](https://mlsanity.com/paper/2412.05237/) (2024)
- [CinePile: A Long Video Question Answering Dataset and Benchmark](https://mlsanity.com/paper/2405.08813/) (2024)
- [HourVideo: 1-Hour Video-Language Understanding](https://mlsanity.com/paper/2411.04998/) (2024)
- [VideoVista: A Versatile Benchmark for Video Understanding and Reasoning](https://mlsanity.com/paper/2406.11303/) (2024)
- [Paxion: Patching Action Knowledge in Video-Language Foundation Models](https://mlsanity.com/paper/2305.10683/) (2023)
- [LvBench: A Benchmark for Long-form Video Understanding with Versatile Multi-modal Question Answering](https://mlsanity.com/paper/2312.04817/) (2023)
- [ALLVB: All-in-One Long Video Understanding Benchmark](https://mlsanity.com/paper/2503.07298/) (2025)
- [LLaVA-Octopus: Unlocking Instruction-Driven Adaptive Projector Fusion for Video Understanding](https://mlsanity.com/paper/2501.05067/) (2025)
- [IV-Bench: A Benchmark for Image-Grounded Video Perception and Reasoning in Multimodal LLMs](https://mlsanity.com/paper/2504.15415/) (2025)
- [HLV-1K: A Large-scale Hour-Long Video Benchmark for Time-Specific Long Video Understanding](https://mlsanity.com/paper/2501.01645/) (2025)

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

- [Video-Thinker: Sparking "Thinking with Videos" via Reinforcement Learning](https://mlsanity.com/paper/2510.23473/) (2025)
- [MT-Video-Bench: A Holistic Video Understanding Benchmark for Evaluating Multimodal LLMs in Multi-Turn Dialogues](https://mlsanity.com/paper/2510.17722/) (2025)
- [VideoTemp-o3: Harmonizing Temporal Grounding and Video Understanding in Agentic Thinking-with-Videos](https://mlsanity.com/paper/2602.07801/) (2026)
- [Thinking with Drafts: Speculative Temporal Reasoning for Efficient Long Video Understanding](https://mlsanity.com/paper/2512.00805/) (2025)
- [TraceAV-Bench: Benchmarking Multi-Hop Trajectory Reasoning over Long Audio-Visual Videos](https://mlsanity.com/paper/2605.07593/) (2026)
- [Seeing the Scene Matters: Revealing Forgetting in Video Understanding Models with a Scene-Aware Long-Video Benchmark](https://mlsanity.com/paper/2603.27259/) (2026)
- [VideoScout: Learning Agentic Active Exploration with Adaptive Reasoning Pacing for Long Video Understanding](https://mlsanity.com/paper/2609.15606/) (2026)
- [VideoEvoHarness: Self Evolving Agent Harness from Unlabeled Videos for Video Understanding](https://mlsanity.com/paper/2610.15389/) (2026)

## Abstract (arXiv)

Although long-video understanding demands that models capture hierarchical temporal information -- from clip (seconds) and shot (tens of seconds) to event (minutes) and story (hours) -- existing benchmarks either neglect this multi-scale design or scatter scale-specific questions across different videos, preventing direct comparison of model performance across timescales on the same content. To address this, we introduce ScaleLong, the first benchmark to disentangle these factors by embedding questions targeting four hierarchical timescales -- clip (seconds), shot (tens of seconds), event (minutes), and story (hours) -- all within the same video content. This within-content multi-timescale questioning design enables direct comparison of model performance across timescales on identical videos. ScaleLong features 269 long videos (avg. 86,min) from 5 main categories and 36 sub-categories, with 4--8 carefully designed questions, including at least one question for each timescale. Evaluating 23 MLLMs reveals a U-shaped performance curve, with higher accuracy at the shortest and longest timescales and a dip at intermediate levels. Furthermore, ablation studies show that increased visual token capacity consistently enhances reasoning across all timescales. ScaleLong offers a fine-grained, multi-timescale benchmark for advancing MLLM capabilities in long-video understanding. The code and dataset are available https://github.com/multimodal-art-projection/ScaleLong.