RECORD
PhysicsGen: Can Generative Models Learn from Images to Predict Complex Physical Relations?
Computer Vision and Pattern Recognition · Submitted 7 Mar 2025
abstract compiled · 2026-09-23 · no language-model review
The authors' main claim While baseline evaluations of different current models show the potential for high speedups (ii), these results also show strong limitations toward the physical correctness (i).
The compiled record
The paper's own sentences, arranged. Each label is drawn from a closed vocabulary; the number beside it is the probability the classifier gave that label. Numbers in the tables are the authors', checked by code. No sentence in this record was written by a model.
Open the compiled record
The abstract, sentence by sentence
The image-to-image translation abilities of generative learning models have recently made significant progress in the estimation of complex (steered) mappings between image distributions. While appearance based tasks like image in-painting or style transfer have been studied at length, we propose to investigate the potential of generative models in the context of physical simulations. Providing a dataset of 300k image-pairs and baseline evaluations for three different physical simulation tasks, we propose a benchmark to investigate the following research questions: i) are generative models able to learn complex physical relations from input-output image pairs? ii) what speedups can be achieved by replacing differential equation based simulations? While baseline evaluations of different current models show the potential for high speedups (ii), these results also show strong limitations toward the physical correctness (i). This underlines the need for new methods to enforce physical correctness. Data, baseline models and evaluation code http://www.physics-gen.org.
background contribution method result interpretation main claim
| # | sentence | role | the authors' own work | states the main conclusion | strength |
|---|---|---|---|---|---|
| s0 | The image-to-image translation abilities of generative learning models have recently made significant progress in the estimation of complex (steered) mappings between image distributions. | background 1.00 | 0.05 | 0.03 | asserted 1.00 |
| s1 | While appearance based tasks like image in-painting or style transfer have been studied at length, we propose to investigate the potential of generative models in the context of physical simulations. | contribution 0.54 | 0.28 | 0.03 | asserted 0.64 |
| s2 | Providing a dataset of 300k image-pairs and baseline evaluations for three different physical simulation tasks, we propose a benchmark to investigate the following research questions: i) are generative models able to learn complex physical relations from input-output image pairs? ii) what speedups can be achieved by replacing differential equation based simulations? | contribution 0.94 | 0.93 | 0.06 | asserted 0.96 |
| s3 | While baseline evaluations of different current models show the potential for high speedups (ii), these results also show strong limitations toward the physical correctness (i). | result 0.90 | 0.89 | 0.49 | asserted 0.99 |
| s4 | This underlines the need for new methods to enforce physical correctness. | interpretation 0.72 | 0.32 | 0.53 | asserted 0.95 |
| s5 | Data, baseline models and evaluation code http://www.physics-gen.org. | contribution 0.88 | 0.29 | 0.02 | unclear 0.65 |
What was examined, and what was not
- arXiv listing abstract v1
- full text
- figures and tables
- references and citing sentences
- relations to other papers
- P abstract
- S sections
- T body sentences
- C citations
- R cross-paper relations
The paper
Original abstract
The image-to-image translation abilities of generative learning models have recently made significant progress in the estimation of complex (steered) mappings between image distributions. While appearance based tasks like image in-painting or style transfer have been studied at length, we propose to investigate the potential of generative models in the context of physical simulations. Providing a dataset of 300k image-pairs and baseline evaluations for three different physical simulation tasks, we propose a benchmark to investigate the following research questions: i) are generative models able to learn complex physical relations from input-output image pairs? ii) what speedups can be achieved by replacing differential equation based simulations? While baseline evaluations of different current models show the potential for high speedups (ii), these results also show strong limitations toward the physical correctness (i). This underlines the need for new methods to enforce physical correctness. Data, baseline models and evaluation code http://www.physics-gen.org.
Human review
References
How the literature uses it
1 arXiv paper cite this in 1 sentence
Papers citing it
Cites it as background 1 paper
-
Physics understanding in image generation evaluates the model's ability to reflect fundamental physical laws and commonsense in visual content [ref] .
Preliminary Explorations with GPT-4o(mni) Native Image Generation · Physics · 0.87
What it cites · 1 sentence without a confident label
Cites as background 2 papers
-
Prior research has demonstrated the effectiveness of embedding physical constraints within AI models to enhance their performance and reliability [ref] .
PhysDiff: Physics-Guided Human Motion Diffusion Model · Related Work. · 0.88
-
Prior research has demonstrated the effectiveness of embedding physical constraints within AI models to enhance their performance and reliability [ref] .
Physically Grounded Vision-Language Models for Robotic Manipulation · Related Work. · 0.88
Discussion
Know something about this paper? Leave a comment, or highlight any passage above to comment on it.