This work investigates a fundamental question: Do Video-Language Models (VidLMs) robustly account for video content, temporal sequence, and motion? Our investigation shows that, surprisingly, they often do not. We introduce REVEAL{}, a diagnostic benchmark that probes fundamental weaknesses of contemporary VidLMs through five controlled stress tests; assessing temporal expectation bias, reliance on language-only shortcuts, video sycophancy, camera motion sensitivity, and robustness to spatiotemporal occlusion. We test leading open- and closed-source VidLMs and find that these models confidently describe reversed scenes as forward, answer questions while neglecting video content, agree with false claims, struggle with basic camera motion, and fail to aggregate temporal information amidst simple spatiotemporal masking. Humans, on the other hand, succeed at these tasks with ease. Alongside our benchmark, we provide a data pipeline that automatically generates diagnostic examples for our stress tests, enabling broader and more scalable evaluation. We will release our benchmark and code to support future research.
@article{arxiv.2602.11244,
title = {Stress Tests REVEAL Fragile Temporal and Visual Grounding in Video-Language Models},
author = {Sethuraman T and Savya Khosla and Aditi Tiwari and Vidya Ganesh and Rakshana Jayaprakash and Aditya Jain and Vignesh Srinivasakumar and Onkar Kishor Susladkar and Srinidhi Sunkara and Aditya Shanmugham and Rakesh Vaideeswaran and Abbaas Alif Mohamed Nishar and Simon Jenni and Derek Hoiem},
journal= {arXiv preprint arXiv:2602.11244},
year = {2026}
}