UniversalVTG: A Universal and Lightweight Foundation Model for Video Temporal Grounding

arXiv:2604.08522v1 Announce Type: new Abstract: Video temporal grounding (VTG) is typically tackled with dataset-specific models that transfer poorly across domains and query styles. Recent efforts to overcome this limitation have adapted large multimodal language models (MLLMs) to VTG, but their high compute cost and limited video context still hinder long-video grounding. We instead scale unified supervision while keeping the model lightweight. We present UniversalVTG, a single VTG model trained with large-scale cross-dataset pretraining. An offline Query Unifier canonicalizes heterogeneous query formats into a shared declarative space, reducing linguistic mismatch and preventing the negative transfer observed under naive joint training. Combined with an efficient grounding head, UniversalVTG scales to long, untrimmed videos. Across diverse benchmarks-GoalStep-StepGrounding, Ego4D-NLQ, TACoS, Charades-STA, and ActivityNet-Captions-one UniversalVTG checkpoint achieves state-of-the-art performance versus dedicated VTG models. Moreover, despite being >100imes smaller than recent MLLM-based approaches, UniversalVTG matches or exceeds their accuracy on multiple benchmarks, offering a practical alternative to parameter-heavy MLLMs.

Related

• Video Parallel Scaling: Aggregating Diverse Frame Subsets for VideoLLMs
• Towards Effective Long Video Understanding of Multimodal Large Language Models via One-shot Clip Retrieval
• Action Without Interaction: Probing the Physical Foundations of Video LMMs via Contact-Release Detection
• CrashSight: A Phase-Aware, Infrastructure-Centric Video Benchmark for Traffic Crash Scene Understanding and Reasoning

Source: arXiv cs.CV | 2026-04-10