Research map / Vision-language and multimodal / Visual question answering
Learning Trimodal Relation for Audio-Visual Question Answering with Missing Modality
Kyu Ri Park, Hong Joo Lee, Jung Uk 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 RMM generator plus AVR diffusion to AVQA networks under missing modality | AVST · MUSIC-AVQA (audio missing) | All Avg accuracy: 36.60 -> 69.71 | Improved |
| add RMM generator plus AVR diffusion | AVST · MUSIC-AVQA (visual missing) | All Avg accuracy: 59.14 -> 67.98 | Improved |
| add RMM generator alone (ablation) | AVST · MUSIC-AVQA (audio missing) | All Avg accuracy: 36.60 -> 60.38 | Improved |
| add AVR diffusion alone (ablation) | AVST · MUSIC-AVQA (audio missing) | All Avg accuracy: 36.60 -> 66.03 | Improved |
| combine RMM generator with AVR diffusion instead of AVR alone (ablation) | AVST · MUSIC-AVQA (audio missing) | All Avg accuracy: 66.03 -> 69.71 | Improved |
| combine RMM generator with AVR diffusion instead of AVR alone (ablation) | AVST · MUSIC-AVQA (visual missing) | All Avg accuracy: 65.62 -> 67.98 | Improved |
| set slot number to 75 instead of 25, 50 or 100 (ablation) | RMM generator · MUSIC-AVQA (audio missing) | All Avg accuracy: 68.85 -> 69.71 | Improved |
| use 10 diffusion time steps instead of 5 or 20 (ablation) | AVR diffusion · MUSIC-AVQA (audio missing) | All Avg accuracy: 68.20 -> 69.71 | Improved |
| proposed method versus existing missing-modality methods on AVST (visual missing) | AVST · MUSIC-AVQA (visual missing) | All Avg accuracy: 67.18 -> 67.98 | Mixed |
| apply method on AVSD where baseline had higher single-column scores | AVSD · AVQA dataset (visual missing) | All Avg accuracy: 45.80 -> 46.35 | Mixed |
Closest papers by meaning
- Audio-Visual Adaptive Fusion Network for Question Answering Based on Contrastive Learning (AAAI 2025 · similarity 0.89)
- AVQACL: A Novel Benchmark for Audio-Visual Question Answering Continual Learning (CVPR 2025 · similarity 0.89)
- Look, Listen, and Answer: Overcoming Biases for Audio-Visual Question Answering (NeurIPS 2024 · similarity 0.88)
- Object-Aware Adaptive-Positivity Learning for Audio-Visual Question Answering (AAAI 2024 · similarity 0.88)
- Question-Aware Gaussian Experts for Audio-Visual Question Answering (CVPR 2025 · similarity 0.86)
- FRAMES-VQA: Benchmarking Fine-Tuning Robustness across Multi-Modal Shifts in Visual Question Answering (CVPR 2025 · similarity 0.86)
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.
