VIDEO | 🇺🇦🇷🇺Footage circulating online is described as showing multiple fires at Russia's Saratov oil refinery following a drone attack. The Rosneft-operated facility has been targeted repeatedly during Ukraine's long-range strike campaign, with previous attacks causing fires and disrupting refining operations. Ukraine's campaign against Russian oil infrastructure is continuing, with refineries remaining among its targets.

VIDEO: Voters turn out in large numbers across the United States for pivotal midterm elections, with long lines reported at polling stations as Americans cast their ballots in closely watched races nationwide.

VIDEO: At China's Hengdian World Studios, tourists can become the stars of their own mini TV dramas, complete with professional cameras, lighting and directors. The experience taps into the country's booming microdrama craze, where episodes can last as little as 30 seconds.

Video from a Los Angeles bus appears to show a black woman repeatedly kicking an elderly passenger in the shin area while wearing boots. The woman allegedly said the elderly passenger had been “disrespecting” her and had run into her. The elderly woman eventually had to be helped off the bus. “That's what yo ugly butt get.”

Video AI turns out to be surprisingly bad at pointing to "when and where," even when it can answer the question itself. Title: OmniCompass: A Holistic Benchmark for Omni-Modal Video Grounding and Understanding URL: 💡 Overview Researchers from Hong Kong Polytechnic University and OPPO Research Institute introduce OmniCompass, a new holistic benchmark for video understanding, evaluating 50 models on temporal/spatial grounding and question answering on the same footing. 🎯 The problem Existing benchmarks cover either vision-only or audio-visual evidence, test either QA or grounding but rarely both, and had no way to check whether a model's predictions actually agree when the target is specified by text alone versus text plus a reference image. 🧪 The method Across 820 videos (80.6 hours, 7 domains, 52 subcategories), each clip gets a matched Text setting and Text+Image setting sharing the same target and ground truth. Two tracks, OmniCompass-VL for vision-language models and OmniCompass-AV for omni-modal models, each evaluate three unified tasks: temporal grounding (TG), spatial grounding (SG), and QA. A new Cross-Specification Consistency (CS) score measures how well predictions agree between the two settings. 📊 Results QA reaches up to 69.1% (VL) and 78.2% (AV), while temporal grounding tops out at just 27.1% (VL) and 40.6% (AV), exposing a huge gap between answering and pointing. The best open-source model also trails the best proprietary model, Gemini-3.8-Flash, by 32.8 points on CS average in the AV track (24.6% vs. 57.4%). #VideoAI# #Benchmarks#