Research map / Vision-language and multimodal / Visual question answering
A Joint Sequence Fusion Model for Video Question Answering and Retrieval
Youngjae Yu, Jongseok Kim, Gunhee Kim
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 Conv-gating (attention) in the convolutional hierarchical decoder (ablation) | JSFusion · LSMDC multiple-choice | accuracy (%): 69.4 -> 73.5 | Improved |
| add Conv-gating in CHD (ablation) | JSFusion · MSR-VTT-MC multiple-choice | accuracy (%): 79.2 -> 83.4 | Improved |
| add Conv-gating in CHD (ablation) | JSFusion · LSMDC movie retrieval | R@10: 28.4 -> 34.1 | Improved |
| add Conv-gating in CHD (ablation) | JSFusion · MSR-VTT-RET retrieval | R@10: 39.5 -> 43.2 | Improved |
| add Conv-gating in CHD (ablation) | JSFusion · LSMDC fill-in-the-blank | accuracy (%): 44.5 -> 45.52 | Improved |
| add audio features (ablation) | JSFusion · LSMDC movie retrieval | R@10: 32.1 -> 34.1 | Improved |
| add audio features (ablation) | JSFusion · MSR-VTT-MC multiple-choice | accuracy (%): 82.9 -> 83.4 | Improved |
| add audio features (ablation) | JSFusion · LSMDC fill-in-the-blank | accuracy (%): 45.26 -> 45.52 | Improved |
| use learnable convolutions in CHD instead of mean pooling (ablation) | JSFusion · LSMDC multiple-choice | accuracy (%): 70.2 -> 73.5 | Improved |
| use learnable convolutions in CHD instead of max pooling (ablation) | JSFusion · MSR-VTT-RET retrieval | R@10: 41.0 -> 43.2 | Improved |
| use BLSTM encoding instead of fully-connected layers (ablation) | JSFusion · LSMDC movie retrieval | R@10: 25.2 -> 34.1 | Improved |
| use BLSTM instead of LSTM encoder (ablation) | JSFusion · LSMDC multiple-choice | accuracy (%): 72.1 -> 73.5 | Improved |
Closest papers by meaning
- LongVideoBench: A Benchmark for Long-context Interleaved Video-Language Understanding (NeurIPS 2024 · similarity 0.89)
- Frame-Voyager: Learning to Query Frames for Video Large Language Models (ICLR 2025 · similarity 0.88)
- MAGNET: A Multi-agent Framework for Finding Audio-Visual Needles by Reasoning over Multi-Video Haystacks (NeurIPS 2025 · similarity 0.87)
- Language-aware Visual Semantic Distillation for Video Question Answering (CVPR 2024 · similarity 0.87)
- Modularized Self-Reflected Video Reasoner for Multimodal LLM with Application to Video Question Answering (ICML 2025 · similarity 0.87)
- Towards Balanced Alignment: Modal-Enhanced Semantic Modeling for Video Moment Retrieval (AAAI 2024 · similarity 0.87)
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.
