REVIEW 11 cited by
Physion: Evaluating Physical Prediction from Vision in Humans and Machines
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
read the original abstract
While current vision algorithms excel at many challenging tasks, it is unclear how well they understand the physical dynamics of real-world environments. Here we introduce Physion, a dataset and benchmark for rigorously evaluating the ability to predict how physical scenarios will evolve over time. Our dataset features realistic simulations of a wide range of physical phenomena, including rigid and soft-body collisions, stable multi-object configurations, rolling, sliding, and projectile motion, thus providing a more comprehensive challenge than previous benchmarks. We used Physion to benchmark a suite of models varying in their architecture, learning objective, input-output structure, and training data. In parallel, we obtained precise measurements of human prediction behavior on the same set of scenarios, allowing us to directly evaluate how well any model could approximate human behavior. We found that vision algorithms that learn object-centric representations generally outperform those that do not, yet still fall far short of human performance. On the other hand, graph neural networks with direct access to physical state information both perform substantially better and make predictions that are more similar to those made by humans. These results suggest that extracting physical representations of scenes is the main bottleneck to achieving human-level and human-like physical understanding in vision algorithms. We have publicly released all data and code to facilitate the use of Physion to benchmark additional models in a fully reproducible manner, enabling systematic evaluation of progress towards vision algorithms that understand physical environments as robustly as people do.
Forward citations
Cited by 11 Pith papers
-
Apple-$\pi$: Benchmarking Thinking with Video Towards Law-Grounded Physical Intelligence
A law-grounded benchmark, Apple-PI, grades video models stage-by-stage on physics reasoning and finds they top out at 0.473, well short of reliable simulation.
-
Vision Language Models Cannot Reason About Physical Transformation
Current VLMs cannot maintain transformation-invariant representations of number, length, volume or size and instead rely on textual invariance priors that reverse on matched non-conserving controls.
-
PhiZero: A World Model Built Around Physical Language
A self-supervised discrete physical-language bottleneck plus a VLM reasoner lets a world model predict state transitions before rendering video, improving physical coherence and enabling zero-shot motion transfer.
-
Thinking in Video: Can Video Generators Really Reason About the Real World?
Video generators show a perception-prediction gap: they can generate plausible continuations while failing explicit visual reasoning tests.
-
Unfolding Spatial Cognition: Evaluating Multimodal Models on Visual Simulations
STARE is a 4K-task benchmark showing multimodal LLMs perform near random chance on multi-step spatial simulation tasks such as cube net folding and tangrams, despite strong 2D transformation results.
-
Seeing is Not Reasoning: MVPBench for Graph-based Evaluation of Multi-path Visual Physical CoT
A new multi-image benchmark and graph-based scoring method show that MLLMs produce weak, poorly-grounded chains of thought on visual physics tasks, and that RL post-training can degrade spatial reasoning.
-
VideoREPA: Learning Physics for Video Generation through Relational Alignment with Foundation Models
VideoREPA adds a token-relation distillation loss that aligns a text-to-video diffusion model's internal features with VideoMAEv2, boosting physical commonsense scores on VideoPhy and VideoPhy2.
-
IMBench: A Benchmark for Intuitive Robotic Manipulation
IMBench is a 35-task robosuite benchmark with a three-stage evaluation showing current VLMs and robot policies fail to convert physical reasoning into executable manipulation.
-
3D and 4D World Modeling: A Survey
A survey that defines 3D/4D world modeling, organizes methods into VideoGen, OccGen, and LiDARGen categories, and compiles datasets, metrics, and benchmark numbers.
-
Pixie: Fast and Generalizable Supervised Learning of 3D Physics from Pixels
A supervised 3D U-Net predicts per-voxel material fields from CLIP feature grids, enabling fast MPM-based animation, but the reported evidence depends on pseudo-labels and a VLM judge from the same model family as the...
-
SlotPi: Physics-informed Object-centric Reasoning Models
SlotPi combines a learned Hamiltonian energy module with spatiotemporal attention to improve object-centric video prediction and visual question answering on several datasets.
Discussion (0). Sign in to comment.