Research map / Video / Video-language understanding
CausalVTG: Towards Robust Video Temporal Grounding via Causal Inference
Qiyi Wang, Senda Chen, Ying Shen
Read the paperCheck your idea against 137,217 papers
AI-extracted summary. Written by AI from the paper's abstract only. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Proposes CausalVTG, a causal-inference approach to video temporal grounding that aims to reduce reliance on dataset-induced co-occurrence biases and improve robustness.
Problem
Video Temporal Grounding (localizing query-relevant segments in untrimmed videos) methods tend to learn superficial co-occurrence patterns induced by dataset biases, which hurts generalization and robustness.
Approach
Applies causal inference to VTG to separate causal from spurious correlations between video content/phrases and event localization, so the model grounds on causal evidence rather than dataset co-occurrences. Only the abstract was available, so the specific mechanism is not described here.
Closest papers by meaning
- Interventional Video Grounding With Dual Contrastive Learning (CVPR 2021 · similarity 0.92)
- ProTeGe: Untrimmed Pretraining for Video Temporal Grounding by Video Temporal Grounding (CVPR 2023 · similarity 0.91)
- TRACE: Temporal Grounding Video LLM via Causal Event Modeling (ICLR 2025 · similarity 0.90)
- VTG-LLM: Integrating Timestamp Knowledge into Video LLMs for Enhanced Video Temporal Grounding (AAAI 2025 · similarity 0.90)
- Empower Words: DualGround for Structured Phrase and Sentence-Level Temporal Grounding (NeurIPS 2025 · similarity 0.90)
- Cross-modal Causal Relation Alignment for Video Question Grounding (CVPR 2025 · similarity 0.90)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
