TSHA: A Benchmark for Visual Language Models in Trustworthy Safety Hazard Assessment Scenarios

TL;DR

TSHA is a comprehensive benchmark designed to evaluate vision-language models in complex, real-world safety hazard assessment scenarios, addressing limitations of existing synthetic and oversimplified datasets.

Contribution

Introduces TSHA, a large, diverse, and challenging benchmark for safety hazard assessment, improving evaluation protocols and model robustness in real-world environments.

Findings

Current VLMs lack robustness in safety hazard assessment.

Training on TSHA improves model performance by up to 18.3 points.

Models trained on TSHA generalize better across different benchmarks.

Abstract

Recent advances in vision-language models (VLMs) have accelerated their application to indoor safety hazards assessment. However, existing benchmarks suffer from three fundamental limitations: (1) heavy reliance on synthetic datasets constructed via simulation software, creating a significant domain gap with real-world environments; (2) oversimplified safety tasks with artificial constraints on hazard and scene types, thereby limiting model generalization; and (3) absence of rigorous evaluation protocols to thoroughly assess model capabilities in complex home safety scenarios. To address these challenges, we introduce TSHA (\textbf{T}rustworthy \textbf{S}afety \textbf{H}azards \textbf{A}ssessment), a comprehensive benchmark comprising 81,809 carefully curated training samples drawn from four complementary sources: existing indoor datasets, internet images, AIGC images, and newly captured images. This benchmark set also includes a highly challenging test set with 1707 samples, comprising not only a carefully selected subset from the training distribution but also newly added videos and panoramic images containing multiple safety hazards, used to evaluate the model's robustness in complex safety scenarios. Extensive experiments on 23 popular VLMs demonstrate that current VLMs lack robust capabilities for safety hazard assessment. Importantly, models trained on the TSHA training set not only achieve a significant performance improvement of up to +18.3 points on the TSHA test set but also exhibit enhanced generalizability across other benchmarks, underscoring the substantial contribution and importance of the TSHA benchmark.