# TimeBlind: A Spatio-Temporal Compositionality Benchmark for Video LLMs

arXiv: https://arxiv.org/abs/2602.00288 · HTML: https://mlsanity.com/paper/2602.00288/

## Key facts

- Cited by 11 papers in the map (data through 14 Oct 2026).
- Builds on 44 of its 58 references that are in the map.
- First posted on arXiv 30 Jan 2026.

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

- [Visual Genome: Connecting Language and Vision Using Crowdsourced Dense Image Annotations](https://mlsanity.com/paper/1602.07332/) (2016)
- [InternVL3.5: Advancing Open-Source Multimodal Models in Versatility, Reasoning, and Efficiency](https://mlsanity.com/paper/2508.18265/) (2025)
- [CLEVR: A Diagnostic Dataset for Compositional Language and Elementary Visual Reasoning](https://mlsanity.com/paper/1612.06890/) (2016)
- [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)
- [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)
- [ActivityNet-QA: A Dataset for Understanding Complex Web Videos via Question Answering](https://mlsanity.com/paper/1906.02467/) (2019)
- [BLINK: Multimodal Large Language Models Can See but Not Perceive](https://mlsanity.com/paper/2404.12390/) (2024)
- [MovieChat: From Dense Token to Sparse Memory for Long Video Understanding](https://mlsanity.com/paper/2307.16449/) (2023)
- [Winoground: Probing Vision and Language Models for Visio-Linguistic Compositionality](https://mlsanity.com/paper/2204.03162/) (2022)
- [TVQA: Localized, Compositional Video Question Answering](https://mlsanity.com/paper/1809.01696/) (2018)
- [LVBench: An Extreme Long Video Understanding Benchmark](https://mlsanity.com/paper/2406.08035/) (2024)
- [TGIF-QA: Toward Spatio-Temporal Reasoning in Visual Question Answering](https://mlsanity.com/paper/1704.04497/) (2017)
- [TempCompass: Do Video LLMs Really Understand Videos?](https://mlsanity.com/paper/2403.00476/) (2024)
- [Perception Test: A Diagnostic Benchmark for Multimodal Video Models](https://mlsanity.com/paper/2305.13786/) (2023)
- [MLVU: Benchmarking Multi-task Long Video Understanding](https://mlsanity.com/paper/2406.04264/) (2024)
- [Video-MMMU: Evaluating Knowledge Acquisition from Multi-Discipline Professional Videos](https://mlsanity.com/paper/2501.13826/) (2025)
- [VideoChat-Flash: Hierarchical Compression for Long-Context Video Modeling](https://mlsanity.com/paper/2501.00574/) (2024)
- [STAR: A Benchmark for Situated Reasoning in Real-World Videos](https://mlsanity.com/paper/2405.09711/) (2024)
- [Video Understanding with Large Language Models: A Survey](https://mlsanity.com/paper/2312.17432/) (2023)
- [MiniCPM-V 4.5: Cooking Efficient MLLMs via Architecture, Data, and Training Recipe](https://mlsanity.com/paper/2509.18154/) (2025)
- [Molmo2: Open Weights and Data for Vision-Language Models with Video Understanding and Grounding](https://mlsanity.com/paper/2601.10611/) (2026)
- [GenAI-Bench: Evaluating and Improving Compositional Text-to-Visual Generation](https://mlsanity.com/paper/2406.13743/) (2024)
- [Cambrian-S: Towards Spatial Supersensing in Video](https://mlsanity.com/paper/2511.04670/) (2025)
- [Revealing Single Frame Bias for Video-and-Language Learning](https://mlsanity.com/paper/2206.03428/) (2022)
- [Video-Holmes: Can MLLM Think Like Holmes for Complex Video Reasoning?](https://mlsanity.com/paper/2505.21374/) (2025)
- [TemporalBench: Benchmarking Fine-grained Temporal Understanding for Multimodal Video Models](https://mlsanity.com/paper/2410.10818/) (2024)
- [NaturalBench: Evaluating Vision-Language Models on Natural Adversarial Samples](https://mlsanity.com/paper/2410.14669/) (2024)

## Built on by (most cited first, 11 of 11)

- [Can Vision-Language Models Solve the Shell Game?](https://mlsanity.com/paper/2603.08436/) (2026)
- [Learning Spatiotemporal Sensitivity in Video LLMs via Counterfactual Reinforcement Learning](https://mlsanity.com/paper/2605.21988/) (2026)
- [SVI-Bench: A Dynamic Microworld for Strategic Video Intelligence](https://mlsanity.com/paper/2605.31529/) (2026)
- [Act More, Decide Less: Skill-Guided Adaptive Action Chunking for Long-Horizon LLM Agents](https://mlsanity.com/paper/2609.02042/) (2026)
- [Tracking the Truth: Object-Centric Spatio-Temporal Monitoring for Video Large Language Models](https://mlsanity.com/paper/2605.08974/) (2026)
- [Plan-and-Verify Video Reward Reasoning with Spatio-Temporal Scene Graph Grounding](https://mlsanity.com/paper/2606.11838/) (2026)
- [Animation2Code: Evaluating Temporal Visual Reasoning in Video-to-Code Generation](https://mlsanity.com/paper/2606.28593/) (2026)
- [What's the Catch? Evaluating Temporal Consistency in Vision-Language Models](https://mlsanity.com/paper/2608.23474/) (2026)
- [MotionBlind: Probing the Illusion of Motion Understanding in Video-LLMs](https://mlsanity.com/paper/2609.09528/) (2026)
- [VidOmni-Bench: A Benchmark for Fine-Grained Video Understanding via Spatio-Temporal Event Verification across Complexity and Duration](https://mlsanity.com/paper/2609.21521/) (2026)
- [STRAND: Benchmarking and Improving Object-Centric Spatio-Temporal Monitoring in Video Large Language Models](https://mlsanity.com/paper/2609.29607/) (2026)

## Abstract (arXiv)

Fine-grained spatio-temporal understanding is essential for video reasoning and embodied AI. Yet, while Multimodal Large Language Models (MLLMs) master static semantics, their grasp of temporal dynamics remains brittle. We present TimeBlind, a diagnostic benchmark for compositional spatio-temporal understanding. Inspired by cognitive science, TimeBlind categorizes fine-grained temporal understanding into three levels: recognizing atomic events, characterizing event properties, and reasoning about event interdependencies. Unlike benchmarks that conflate recognition with temporal reasoning, TimeBlind leverages a minimal-pairs paradigm: video pairs share identical static visual content but differ solely in temporal structure, utilizing complementary questions to neutralize language priors. Evaluating over 20 state-of-the-art MLLMs (e.g., GPT-5, Gemini 3 Pro) on 600 curated instances (2400 video-question pairs), reveals that the Instance Accuracy (correctly distinguishing both videos in a pair) of the best performing MLLM is only 48.2%, far below the human performance (98.2%). These results demonstrate that even frontier models rely heavily on static visual shortcuts rather than genuine temporal logic, positioning TimeBlind as a vital diagnostic tool for next-generation video understanding. Dataset and code are available at https://baiqi-li.github.io/timeblind_project/ .