Grounding in the Dark: Text-Only Training for Zero-Shot Video Temporal Localization
cs.CV, cs.AI
Submitted: 2026-11-04
Updated: 2026-11-05
Terminology
Sources
- MarkIt: Training-Free Visual Markers for Precise Video Temporal Grounding
- Watching Synthetic Videos: Aligning Cross-modal Representations with Visual Synthesis for Zero-shot Video Captioning
- Your VLM Already Knows When: Training-Free Temporal Grounding by Asking Yes or No
- How Should Video LLMs Output Time? An Analysis of Efficient Temporal Grounding Paradigms
- From Accuracy to Visual Dependence: Auditing and Filtering Modality Collapse in Traffic VideoQA
- Mitigating Modality and Language-Style Gaps for Zero-Shot Video Moment Retrieval
- TAG: A Simple Yet Effective Temporal-Aware Approach for Zero-Shot Video Temporal Grounding
- REZE: Recognition-Based Zero-Shot Extraction for Video Temporal Grounding
- VideoChat: Chat-Centric Video Understanding
- TempSamp-R1: Effective Temporal Sampling with Reinforcement Fine-Tuning for Video LLMs
- LAION-400M: Open Dataset of CLIP-Filtered 400 Million Image-Text Pairs
- HawkEye: Training Video-Text LLMs for Grounding Text in Videos
- Qwen3 Technical Report
- Anisotropic Modality Align
- TD-VAD: Breaking Visual Dependence in Video Anomaly Detection with Text-Driven Learning
- DART: Difficulty-Adaptive Routing for Zero-Shot Video Temporal Grounding
Related papers
- TEA: Text Encoder Alignment for Robust Concept Erasure in Text-to-Image Models
- Do Egocentric Video-Language Models Capture Both Hand- and Object-Centric Cues?
- Aura: Consistent Multi-Subject Video Generation via VLM-Grounded Semantic Alignment
- PVRF: All-in-one Adverse Weather Removal via Prior-modulated and Velocity-constrained Rectified Flow
- DLWM: Diverse Latent World Models for Efficient Multimodal Reasoning
- Out of Sight, Still in Mind: Token Compression for Omni-LLMs