paper-with-me

홈 › Papers

PHYRE: A New Benchmark for Physical Reasoning

2019-08-15 · NeurIPS 2019 12 · Anton Bakhtin, Laurens van der Maaten, Justin Johnson, Laura Gustafson, Ross Girshick

Understanding and reasoning about physics is an important ability of intelligent agents. We develop the PHYRE benchmark for physical reasoning that contains a set of simple classical mechanics puzzles in a 2D physical environment. The benchmark is designed to encourage the development of learning algorithms that are sample-efficient and generalize well across puzzles. We test several modern learning algorithms on PHYRE and find that these algorithms fall short in solving the puzzles efficiently. We expect that PHYRE will encourage the development of novel sample-efficient agents that learn efficient but useful models of physics. For code and to play PHYRE for yourself, please visit https://player.phyre.ai.

📄 PDF Abstract BibTeX arXiv:1908.05656

Code (2)

facebookresearch/phyre 공식 구현
ndrwmlnk/awesome-benchmarks-for-physical-reasoning-ai

Tasks

Visual Reasoning

Similar Papers 제목 키워드 기반

InPhyRe Discovers: Large Multimodal Models Struggle in Inductive Physical Reasoning

2025-09-12 · Gautam Sreekumar, Vishnu Naresh Boddeti arxiv

Large multimodal models (LMMs) encode physical laws observed during training, such as momentum conservation, as parametric knowledge. It allows LMMs to answer physical reasoning queries, such as the outcome of a potentia…

Visual Question Answering

Forward Prediction for Physical Reasoning

2020-06-18 · Rohit Girdhar, Laura Gustafson, Aaron Adcock, Laurens van der Maaten

Physical reasoning requires forward prediction: the ability to forecast what will happen next given some initial world state. We study the performance of state-of-the-art forward-prediction models in the complex physical…

PredictionVisual Reasoning

I-PHYRE: Interactive Physical Reasoning

2023-12-04 · Shiqian Li, Kewen Wu, Chi Zhang, Yixin Zhu

Current evaluation protocols predominantly assess physical reasoning in stationary scenes, creating a gap in evaluating agents' abilities to interact with dynamic events. While contemporary methods allow agents to modify…

Zero-shot Generalization

Causal-PIK: Causality-based Physical Reasoning with a Physics-Informed Kernel

2025-05-28 · Carlota Parés-Morlans, Michelle Yi, Claire Chen, Sarah A. Wu 외

Tasks that involve complex interactions between objects with unknown dynamics make planning before execution difficult. These tasks require agents to iteratively improve their actions after actively exploring causes and …

Bayesian Optimization

PhyRPR: Training-Free Physics-Constrained Video Generation

2026-01-14 · Yibo Zhao, Hengjia Li, Xiaofei He, Boxi Wu arxiv

Recent diffusion-based video generation models can synthesize visually plausible videos, yet they often struggle to satisfy physical constraints. A key reason is that most existing approaches remain single-stage: they en…

Video Generation