Research map / Vision-language and multimodal / Visual question answering
What You Ask is What You Ground: Bridging Question Intent to Temporal Evidence for Grounded VideoQA
Jinhwan Seo, Kyu Han, Jumin Lee, Junhyug Noh, Sung-eui Yoon
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add communication tokens (Linguistic Transfer and Visual Refinement) (ablation) | GroundFormer · NExT-GQA | Acc@GQA: 14.8 -> 18.6 | Improved |
| add question-type loss L_type (ablation) | GroundFormer · NExT-GQA | Acc@GQA: 18.6 -> 19.0 | Improved |
| add hierarchical contrastive loss L_mcl (ablation) | GroundFormer · NExT-GQA | Acc@GQA: 19.0 -> 20.1 | Improved |
| add Gaussian smoothing (ablation) | GroundFormer · NExT-GQA | Acc@GQA: 20.1 -> 20.8 | Improved |
| add Gaussian smoothing (ablation) | GroundFormer · NExT-GQA | mIoU: 17.2 -> 16.5 | Hurt |
| add MIL cross-attention (ablation) | GroundFormer · NExT-GQA | Acc@GQA: 20.8 -> 21.5 | Improved |
| add two-pass training with question pass (ablation) | GroundFormer · NExT-GQA | Acc@GQA: 15.7 -> 21.5 | Improved |
| add communication tokens (ablation) | GroundFormer · NExT-GQA | PIoU: 57.4 -> 43.5 | Improved |
| add contrastive loss L_mcl (ablation) | GroundFormer · NExT-GQA | PIoU: 37.9 -> 29.4 | Improved |
| add Gaussian smoothing (ablation) | GroundFormer · NExT-GQA | PIoU: 29.4 -> 29.4 | No clear effect |
| add MIL without two-pass training (ablation) | GroundFormer · NExT-GQA | PIoU: 51.4 / 0.05 | No clear effect |
| use MIL cross-attention instead of WSTAN objective (ablation) | GroundFormer · NExT-GQA | Acc@GQA: 20.0 -> 21.5 | Improved |
Closest papers by meaning
- Universal Video Temporal Grounding with Generative Multi-modal Large Language Models (NeurIPS 2025 · similarity 0.91)
- Can I Trust Your Answer? Visually Grounded Video Question Answering (CVPR 2024 · similarity 0.90)
- Searching Videos as Trees: Self-Correcting Agents for Grounded Long Video QA (NeurIPS 2026 · similarity 0.90)
- Grounded Multi-Hop VideoQA in Long-Form Egocentric Videos (AAAI 2025 · similarity 0.90)
- Grounded Question-Answering in Long Egocentric Videos (CVPR 2024 · similarity 0.89)
- Modularized Self-Reflected Video Reasoner for Multimodal LLM with Application to Video Question Answering (ICML 2025 · similarity 0.89)
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.
