Physics-IQ Verified

Hugging Face Daily Papers Papers

Summary

This paper presents a systematic audit of the Physics-IQ benchmark for evaluating physical understanding in video generative models, proposing improvements to prompts and scoring to enhance reliability.

Video generative models ( VGMs) have become a new frontier that can be used not just for video generation but for a multitude of downstream tasks, including world modeling. To advance these tasks, a good video model must understand the physical reality of the world. Evaluating this understanding is an emerging field and has led to the Physics-IQ benchmark, which quantifies this explicitly by comparing model-generated videos to real-world videos of physical experiments. In this work, we present a systematic audit of the Physics-IQ benchmark, expose shortcomings and propose three solutions that sharpen how we can measure physical understanding of VGMs. Specifically, we improve prompt and ground-truth quality to reduce the influence of confounding factors and further introduce a sample-level scoring system that weights each sample and metric equally. Our resulting benchmark, Physics-IQ Verified, refines 57.6\% of all samples and improves over 34.8\% of prompts. In a comparison study using six image-to-video generative models, we observe moderate but meaningful ranking changes (Kendall's τ= 0.46). We hope Physics-IQ Verified advances the community by providing a more reliable signal toward physically accurate VGMs. The code for the benchmark can be accessed at https://github.com/google-deepmind/physics-iq-benchmark
Original Article
View Cached Full Text

Cached at: 06/18/26, 03:55 AM

Paper page - Physics-IQ Verified

Source: https://huggingface.co/papers/2606.18943

Abstract

A systematic evaluation of the Physics-IQ benchmark reveals limitations in measuring physical understanding of video generative models, leading to improvements in prompt quality and sample-level scoring that enhance reliability for assessing physically accurate video generation.

Video generative models( VGMs) have become a new frontier that can be used not just for video generation but for a multitude of downstream tasks, includingworld modeling. To advance these tasks, a good video model must understand the physical reality of the world. Evaluating this understanding is an emerging field and has led to thePhysics-IQ benchmark, which quantifies this explicitly by comparing model-generated videos to real-world videos of physical experiments. In this work, we present a systematic audit of thePhysics-IQ benchmark, expose shortcomings and propose three solutions that sharpen how we can measurephysical understandingof VGMs. Specifically, we improve prompt and ground-truth quality to reduce the influence of confounding factors and further introduce asample-level scoringsystem that weights each sample and metric equally. Our resulting benchmark, Physics-IQ Verified, refines 57.6\% of all samples and improves over 34.8\% of prompts. In a comparison study using six image-to-video generative models, we observe moderate but meaningful ranking changes (Kendall’s τ= 0.46). We hope Physics-IQ Verified advances the community by providing a more reliable signal toward physically accurate VGMs. The code for the benchmark can be accessed at https://github.com/google-deepmind/physics-iq-benchmark

View arXiv pageView PDFAdd to collection

Get this paper in your agent:

hf papers read 2606\.18943

Don’t have the latest CLI?curl \-LsSf https://hf\.co/cli/install\.sh \| bash

Models citing this paper0

No model linking this paper

Cite arxiv.org/abs/2606.18943 in a model README.md to link it from this page.

Datasets citing this paper0

No dataset linking this paper

Cite arxiv.org/abs/2606.18943 in a dataset README.md to link it from this page.

Spaces citing this paper0

No Space linking this paper

Cite arxiv.org/abs/2606.18943 in a Space README.md to link it from this page.

Collections including this paper0

No Collection including this paper

Add this paper to acollectionto link it from this page.

Similar Articles

Physics-R1: An Audited Olympiad Corpus and Recipe for Visual Physics Reasoning

arXiv cs.CL

This paper audits multimodal physics evaluation pipelines, revealing issues like train-eval contamination, translation drift, and MCQ saturation. It releases new datasets (PhysCorp-A, PhysR1Corp, PhysOlym-A) and a training recipe (Physics-R1) that significantly improves performance on held-out olympiad problems.

Apple-π: Benchmarking Thinking with Video Towards Law-Grounded Physical Intelligence

Hugging Face Daily Papers

Apple-π is a benchmark that evaluates video generation models on their ability to reason about physical laws through a three-stage protocol: perception, formulation, deduction. It includes 400 videos covering classical mechanics tasks and reveals current models fall short of reliable law-grounded world simulation.

Physics-Audited Agentic Discovery in Scientific Machine Learning

arXiv cs.AI

Introduces Physics-Audited Agentic SciML (PA-SciML), a verification-first workflow where LLM agents discover surrogate models and validate them against physics requirements such as boundary conditions and causality, not just error metrics. Numerical examples show improved trustworthiness over error-only baselines.