datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
physics-course-vidsphysics
CAMEL: Communicative Agents for “Mind” Exploration of Large Scale Language Model Society
Github: https://github.com/lightaime/camel
Website: https://www.camel-ai.org/
Arxiv Paper: https://arxiv.org/abs/2303.17760
Dataset Summary
Physics dataset is composed of 20K problem-solution pairs obtained using gpt-4. The dataset problem-solutions pairs generating from 25 physics topics, 25 subtopics for each topic and 32 problems for each "topic,subtopic" pairs.
We… See the full description on the dataset page: https://huggingface.co/datasets/camel-ai/physics.physics-corpus
Physics Corpus — konsman/physics-corpus
arXiv physics papers exported from a PostgreSQL mirror of the Kaggle arXiv dataset,
structured for ontology extraction and downstream NLP pipelines.
Configuration: quantum-physics
arXiv categories included: quant-ph, hep-th, gr-qc
Schema
Field
Type
Description
paper_id
string
arxiv:<id>v<n> — stable across pipeline runs
arxiv_id
string
Base arXiv ID without version suffix
arxiv_version
int32
Version number… See the full description on the dataset page: https://huggingface.co/datasets/konsman/physics-corpus.vidore_v3_physicsViDoRe V3 : Physics
This dataset, Physics, is a corpus of course slides on bachelor level physics lectures, intended for complex visual understanding tasks. It is one of the 10 corpora comprising the ViDoRe v3 Benchmark.
About ViDoRe v3
ViDoRe V3 is our latest benchmark for RAG evaluation on visually-rich documents from real-world applications. It features 10 datasets with, in total, 26,000 pages and 3099 queries, translated into 6 languages. Each query comes with human-verified… See the full description on the dataset page: https://huggingface.co/datasets/vidore/vidore_v3_physics.morpheus-real-world
Morpheus — Real-World Physics Videos
Real-world reference footage for Morpheus, a benchmark that tests whether
video generative models (Wan, CogVideo, LTX-Video, COSMOS-predict1/2,
Pyramid-Flow, Veo3, Kling-Turbo, ...) obey Newtonian mechanics. Object
trajectories are extracted via SAM2 tracking and tested against physical laws
(energy/momentum conservation, equations of motion) rather than pixel-matched
to a single "correct" video.
This repo contains the filmed real-world… See the full description on the dataset page: https://huggingface.co/datasets/physics-from-video/morpheus-real-world.meta13sphere_IRS_DCE_Topological_Dynamics__Boundary_Dissolution_Physics
Resonance Resonance / IRS-DCE
MASTER README (FULL EXTENDED VERSION)
If you need the other data or pdf check on [https://huggingface.co/datasets/meta13sphere/phaseShift_shell_result_pdf]
[2026-09-25 Update]
The Geometry of IRS-DCE Boundary Dissolution v1.0 is now available in Korean and English.
Resolution-Dependent Component Decomposition and Dynamic Rearrangement
This release connects the existing IRS-DCE and BBRCM research to mathematical analysis and recorded… See the full description on the dataset page: https://huggingface.co/datasets/meta13sphere/meta13sphere_IRS_DCE_Topological_Dynamics__Boundary_Dissolution_Physics.cqadupstack-physics
CQADupstackPhysicsRetrieval
An MTEB dataset
Massive Text Embedding Benchmark
CQADupStack: A Benchmark Data Set for Community Question-Answering Research
Task category
t2t
Domains
Written, Academic, Non-fiction
Referencehttp://nlp.cis.unimelb.edu.au/resources/cqadupstack/
How to evaluate on this task
You can evaluate an embedding model on this dataset using the following code:
import mteb
task = mteb.get_tasks(["CQADupstackPhysicsRetrieval"])
evaluator… See the full description on the dataset page: https://huggingface.co/datasets/mteb/cqadupstack-physics.agieval-gaokao-physics
Dataset Card for "agieval-gaokao-physics"
Dataset taken from https://github.com/microsoft/AGIEval and processed as in that repo, following dmayhem93/agieval-* datasets on the HF hub.
This dataset contains the contents of the Gaokao Physics subtask of AGIEval, as accessed in https://github.com/ruixiangcui/AGIEval/commit/5c77d073fda993f1652eaae3cf5d04cc5fd21d40 .
Citation:
@misc{zhong2023agieval,
title={AGIEval: A Human-Centric Benchmark for Evaluating Foundation Models}… See the full description on the dataset page: https://huggingface.co/datasets/hails/agieval-gaokao-physics.physicsiq-candidatesDownstream_Physics_Simulation
GeoPT
Project Page | Paper | GitHub
This repository contains the physics simulation data for the paper GeoPT: Scaling Physics Simulation via Lifted Geometric Pre-Training.
GeoPT is a unified model pre-trained on large-scale geometric data for general physics simulation, unlocking a scalable path for neural simulation.
Overview
GeoPT is evaluated on the following five simulation tasks.
Dataset
Mesh Size
Variable
Training
Test
Total Size
Source
DrivAerML… See the full description on the dataset page: https://huggingface.co/datasets/GeoPT/Downstream_Physics_Simulation.Physics-IQ-Verified
Physics-IQ Verified Dataset
This repository hosts the Physics-IQ Verified benchmark data for evaluating physical understanding in generative video models.
Physics-IQ Verified is derived from the original Physics-IQ benchmark dataset.
Original Physics-IQ
Paper: Do generative video models understand physical principles?
Repository: Code | Dataset in Google Cloud
Physics-IQ Verified (Recommended)
Paper: Physics-IQ Verified
Repository: Code | Dataset: Here in this repo :)
We… See the full description on the dataset page: https://huggingface.co/datasets/Anates-Labs-Research/Physics-IQ-Verified.physics-r1-eval-outputsworld-model-physics-human-preference-283k
Rapidata Physics Benchmark
Built by Rapidata.
Do video and world models understand physics? We gave 25 video- and world models the same
real-world starting frame and scene description from Physics-IQ and asked
them to predict what happens next. ~283,000 human votes, collected with the
Rapidata Python SDK, decided which continuation is more realistic — with the
real recording competing as a hidden 26th participant.
Each row is a head-to-head matchup between two participants on… See the full description on the dataset page: https://huggingface.co/datasets/Rapidata/world-model-physics-human-preference-283k.physics-stackexchangevideo-gen-physics-gallery
Paper material (assets/drive)
Two generated galleries plus a figure asset pack. All reproducible from the repo —
don't hand-edit them, re-run the generator.
gallery
what it shows
generator
real_videos/
the input side of the benchmark: real GT episodes, 5 diverse tasks per (embodiment x view-track x markovian/non-markovian) cell
scripts/make_real_video_gallery.py
method_comparison/
the output side: ONE episode rendered by every acceleration method, so a single dir is… See the full description on the dataset page: https://huggingface.co/datasets/doanh25032004/video-gen-physics-gallery.wan22-physics-videos
Wan2.2 Physics Video Generation Dataset
A dataset of AI-generated physics simulation videos with intermediate denoising latent tensors, created using the Wan2.2 T2V-A14B text-to-video model.
Dataset Summary
480 videos (.mp4) generated from 60 unique physics prompts (8 random seeds each)
Intermediate latent tensors (.pt) saved at every 2 denoising steps (10 snapshots per video)
Physics categories: ball bouncing, pendulum motion, objects sliding on inclined planes… See the full description on the dataset page: https://huggingface.co/datasets/hivamoh/wan22-physics-videos.agieval-gaokao-physics
Dataset Card for "agieval-gaokao-physics"
Dataset taken from https://github.com/microsoft/AGIEval and processed as in that repo.
MIT License
Copyright (c) Microsoft Corporation.
Permission is hereby granted, free of charge, to any person obtaining a copy
of this software and associated documentation files (the "Software"), to deal
in the Software without restriction, including without limitation the rights
to use, copy, modify, merge, publish, distribute, sublicense, and/or sell… See the full description on the dataset page: https://huggingface.co/datasets/dmayhem93/agieval-gaokao-physics.Physics-Informed-Deformable-Gaussian-SplattingRobot_physics_interaction_with_Human_linguistic
Dual-Camera Speech and Interaction Dataset
This directory contains 300 recording sessions. Each session combines first-person video, third-person video, human speech and a text label derived from the read-aloud script.
For the research context behind the dataset, see PROJECT_README.md.
Directory structure
ACTION_NUMBER/
├── first_person_view.mp4
├── third_person_view.mp4
├── audio.wav
├── text.txt
├── label.json
├── first_person_view_camera/
│ ├──… See the full description on the dataset page: https://huggingface.co/datasets/iciclab/Robot_physics_interaction_with_Human_linguistic.inverse-physics-warp-dataset
Inverse Physics Warp Dataset
Synthetic particle trajectories generated with Warp MPM for constitutive learning
and position-only material inference. The versioned recipe and archive inventory
are in recipe.json and manifest.json. A release is complete only when every
archive listed in the manifest is available.
Version 1
Stage 1 contains 800 homogeneous trajectories: five material families (elastic,
Newtonian, non-Newtonian, plasticine, sand), 40 parameter samples… See the full description on the dataset page: https://huggingface.co/datasets/ZhewenZheng/inverse-physics-warp-dataset.vidore_v3_physics_mteb_format
Vidore3PhysicsRetrieval
An MTEB dataset
Massive Text Embedding Benchmark
Retrieve associated pages according to questions.
Task category
t2i
Domains
Academic
Reference
https://huggingface.co/blog/QuentinJG/introducing-vidore-v3
Source datasets:
vidore/vidore_v3_physics
How to evaluate on this task
You can evaluate an embedding model on this dataset using the following code:
import mteb
task = mteb.get_task("Vidore3PhysicsRetrieval")
evaluator =… See the full description on the dataset page: https://huggingface.co/datasets/vidore/vidore_v3_physics_mteb_format.EWT-Vacuum-Lattice-Unified-Physics
Enhanced EWT: BCC Vacuum Lattice Framework
This dataset contains the full theoretical and computational framework for the Enhanced Energy Wave Theory (EWT). It provides a deterministic derivation of fundamental physical constants using a Body-Centered Cubic (BCC) vacuum lattice geometry.
🎯 Key Breakthroughs
N_geometric = 8π⁴: Derivation of vacuum stiffness ($N$) from BCC lattice geometry.
G-constant Precision: Achieved absolute delta of 1.29e-25 vs CODATA 2022.… See the full description on the dataset page: https://huggingface.co/datasets/luksmol/EWT-Vacuum-Lattice-Unified-Physics.PhysicsNeMo-Datacenter-CFD
Overview
This asset provides the dataset and reference OpenFOAM configuration to be used for the Datacenter CFD use case here.
The dataset contains normalized OpenFOAM simulations for various configurations of a typical hot aisle in a datacenter. The dataset includes variation for the number of IT racks and hot-aisle geometry parameters like length, height and width of the hot aisle. These configurations are solved with OpenFOAM assuming maximum flow rate and rack exit temperature… See the full description on the dataset page: https://huggingface.co/datasets/nvidia/PhysicsNeMo-Datacenter-CFD.dataset-CoT-Physics-2254Downstream_Physics_Simulation
GeoPT
Project Page | Paper | GitHub
This repository contains the physics simulation data for the paper GeoPT: Scaling Physics Simulation via Lifted Geometric Pre-Training.
GeoPT is a unified model pre-trained on large-scale geometric data for general physics simulation, unlocking a scalable path for neural simulation.
Overview
GeoPT is evaluated on the following five simulation tasks.
Dataset
Mesh Size
Variable
Training
Test
Total Size
Source… See the full description on the dataset page: https://huggingface.co/datasets/introvoyz041/Downstream_Physics_Simulation.video_gen_physics_real_videophysicsgen
PhysicsGen: Can Generative Models Learn from Images to Predict Complex Physical Relations?
Paper
Accepted at IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) 2025.
Preprint is available here: https://arxiv.org/abs/2503.05333
Website: https://www.physics-gen.org/
Github: https://github.com/physicsgen/physicsgen
Overview
PhysicsGen is a synthetic dataset collection generated via simulation for physical guided generative modeling, focusing… See the full description on the dataset page: https://huggingface.co/datasets/mspitzna/physicsgen.Physicsvideo_gen_physicsAtom-Harness-Seedance-2.5-Physics-IQ-Verified
Atom Harness (Seedance 2.5) — Physics-IQ Verified, multiframe (v2v), best-practice prompts (bpp)
Physics-IQ Verified score: 56.6 — 198/198 cases, single sample per case (no best-of-N, no selection, no reranking), one run (seed 7).
This dataset is the evidence bundle for a Physics-IQ Verified leaderboard entry: all 198 generated videos, per-case component scores, and the run summary. Scores were computed with the official evaluator (physiq/run_physics_iq.py, verified ground truth… See the full description on the dataset page: https://huggingface.co/datasets/ssaroya/Atom-Harness-Seedance-2.5-Physics-IQ-Verified.