Dataset · Hugging Face

Navinvue/HomeSafeBench

Name: HomeSafeBench
Creator: Navinvue
License: cc-by-4.0
Keywords: language:en, license:cc-by-4.0, size_categories:10K<n<100K, modality:image, modality:text, arxiv:2509.23690, region:us, embodied-ai, vision-language, home-safety, virtualhome

HomeSafeBench This repository contains the official data release for the paper HomeSafeBench: Benchmarking Embodied Vision-Language Models in Free-Exploration Home Safety Inspection. HomeSafeBench is a benchmark for free-exploration home safety inspection with embodied vision-language models, built on VirtualHome. Dataset Contents data/train/: 3,400 training tasks. data/test/: 1,000 human-validated evaluation tasks. The CueBack data contains 3,158 aligned… See the full description on the dataset page: https://huggingface.co/datasets/Navinvue/HomeSafeBench.

View on Hugging FaceReport an issue

Author / organization	Navinvue
License (as declared)	cc-by-4.0
Downloads	425 (as reported by Hugging Face)
Likes	0
Created	7 August 2026
Last updated at source	11 August 2026

embodied-aivision-languagehome-safetyvirtualhome

Embed this entry

<iframe src="https://silytics.si/embed/item/7a8b2f5a-c81a-4f97-ae4e-8db249af4794" width="420" height="190" style="border:0" loading="lazy" title="Silytics entry"></iframe>

Free, no tracking. More about feeds and embeds.

Silytics shows public metadata from Hugging Face and links to the original. Information may be incomplete or out of date. Last refreshed 16 October 2026. Licenses and terms of the original apply to the work itself. Not financial, legal or professional advice.