mlsanityData to 7 Oct 2026

IntPhys: A Framework and Benchmark for Visual Intuitive Physics Reasoning

arXiv 1803.07616 ↗First posted 20 Mar 2018cs.AI cs.CVcited by 72

Not on the map. Why?

Abstract (arXiv)

In order to reach human performance on complexvisual tasks, artificial systems need to incorporate a sig-nificant amount of understanding of the world in termsof macroscopic objects, movements, forces, etc. Inspiredby work on intuitive physics in infants, we propose anevaluation benchmark which diagnoses how much a givensystem understands about physics by testing whether itcan tell apart well matched videos of possible versusimpossible events constructed with a game engine. Thetest requires systems to compute a physical plausibilityscore over an entire video. It is free of bias and cantest a range of basic physical reasoning concepts. Wethen describe two Deep Neural Networks systems aimedat learning intuitive physics in an unsupervised way,using only physically possible videos. The systems aretrained with a future semantic mask prediction objectiveand tested on the possible versus impossible discrimi-nation task. The analysis of their results compared tohuman data gives novel insights in the potentials andlimitations of next frame prediction architectures.

Key facts. Cited by 72 papers in the map (data through 7 Oct 2026). First posted on arXiv 20 Mar 2018.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 11; 2022: 5; 2023: 15; 2024: 3; 2025: 15; 2026: 23.

72 papers.

Sort:

By area:

  1. 2023Perception Test: A Diagnostic Benchmark for Multimodal Video Modelscited by 310
  2. 2025Cosmos-Reason1: From Physical Common Sense To Embodied Reasoningcited by 159
  3. 2021Physion: Evaluating Physical Prediction from Vision in Humans and Machinescited by 145
  4. 2021Grounding Physical Concepts of Objects and Events Through Dynamic Visual Reasoningcited by 68
  5. 2022ComPhy: Compositional Physical Reasoning of Objects and Events from Videoscited by 61
  6. 2025Embodied AI Agents: Modeling the Worldcited by 57
  7. 2025Back to the Features: DINO as a Foundation for Video World Modelscited by 56
  8. 2021Dynamic Visual Reasoning by Learning Differentiable Physics Models from Video and Languagecited by 55
  9. 2023Physion++: Evaluating Physical Scene Understanding that Requires Online Inference of Different Physical Propertiescited by 53
  10. 2021AGENT: A Benchmark for Core Psychological Reasoningcited by 42
  11. 2026Learning Latent Action World Models In The Wildcited by 42
  12. 2023Benchmarks for Automated Commonsense Reasoning: A Surveycited by 40
38 more
  1. 2025CausalVQA: A Physically Grounded Causal Reasoning Benchmark for Video Modelscited by 35
  2. 2022A Review of Emerging Research Directions in Abstract Visual Reasoningcited by 33
  3. 2025A Shortcut-aware Video-QA Benchmark for Physical Understanding via Minimal Video Pairscited by 30
  4. 2023GRASP: A novel benchmark for evaluating language GRounding And Situated Physics understanding in multimodal language modelscited by 28
  5. 2024ContPhy: Continuum Physical Concept Learning and Reasoning from Videoscited by 25
  6. 2025Generative Physical AI in Vision: A Surveycited by 24
  7. 2021Capturing the objects of vision with neural networkscited by 21
  8. 2022A Survey on Machine Learning Approaches for Modelling Intuitive Physicscited by 20
  9. 2026Interpreting Physics in Video World Modelscited by 18
  10. 2023Learning 3D Particle-based Simulators from RGB-D Videoscited by 17
  11. 2022CRIPP-VQA: Counterfactual Reasoning about Implicit Physical Properties via Video Question Answeringcited by 13
  12. 2023Understanding Physical Dynamics with Counterfactual World Modelingcited by 11
  13. 2021Phy-Q as a measure for physical reasoning intelligencecited by 10
  14. 2021A Benchmark for Modeling Violation-of-Expectation in Physical Reasoning Across Event Categoriescited by 8
  15. 2023Neural Reasoning About Agents' Goals, Preferences, and Actionscited by 8
  16. 2023Benchmarks for Physical Reasoning AIcited by 8
  17. 2026MiraBench: Evaluating Action-Conditioned Reliability in Robotic World Modelscited by 8
  18. 2023A Survey of Methods, Challenges and Perspectives in Causalitycited by 6
  19. 2023Imagine the Unseen World: A Benchmark for Systematic Generalization in Visual World Modelscited by 6
  20. 2025SlotPi: Physics-informed Object-centric Reasoning Modelscited by 5
  21. 2025VKnowU: Evaluating Visual Knowledge Understanding in Multimodal LLMscited by 5
  22. 2021AVoE: A Synthetic 3D Dataset on Understanding Violation of Expectation for Artificial Cognitioncited by 4
  23. 2021OPEn: An Open-ended Physics Environment for Learning Without a Taskcited by 4
  24. 2026Data Pyramid for Embodied Manipulation: A Surveycited by 4
  25. 2023Visual Transformation Tellingcited by 3
  26. 2023X-VoE: Measuring eXplanatory Violation of Expectation in Physical Eventscited by 3
  27. 2023The Acquisition of Physical Knowledge in Generative Neural Networkscited by 3
  28. 2024Object Dynamics Modeling with Hierarchical Point Cloud-based Representationscited by 3
  29. 2025Objects in Generated Videos Are Slower Than They Appear: Models Suffer Sub-Earth Gravity and Don't Know Galileo's Principle...for nowcited by 3
  30. 2026YoCausal: How Far is Video Generation from World Model? A Causality Perspectivecited by 3
  31. 2026World Models: A Comprehensive Survey of Architectures, Methodologies, Reasoning Paradigms, and Applicationscited by 3
  32. 2023NovPhy: A Testbed for Physical Reasoning in Open-world Environmentscited by 2
  33. 2025Causal-PIK: Causality-based Physical Reasoning with a Physics-Informed Kernelcited by 2
  34. 2025Pixels to Principles: Probing Intuitive Physics Understanding in Multimodal Language Modelscited by 2
  35. 2025InPhyRe Discovers: Large Multimodal Models Struggle in Inductive Physical Reasoningcited by 2
  36. 2026PhysEditWorld: A Large-Scale Dataset Toward Physics-Editable World Modelscited by 2
  37. 2021Hi-Phy: A Benchmark for Hierarchical Physical Reasoningcited by 1
  38. 2026Emergent Compositional Communication for Latent World Propertiescited by 1

50 most cited of 72 All 72 (JSON)