Research map / Vision-language and multimodal / Cross-modal learning
Weakly Paired Associative Learning for Sound and Image Representations via Bimodal Associative Memory
Sangmin Lee, Hyung-Il Kim, Yong Man Ro
Check 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 unpaired sound data from Kinetics-400 (ablation) | BMA-Memory weakly paired associative learning · ACIVW image classification | Top-1 accuracy (linear evaluation): 0.745 -> 0.778 | Improved |
| Add unpaired image data from Kinetics-400 (ablation) | BMA-Memory weakly paired associative learning · ACIVW sound classification | Top-1 accuracy (linear evaluation): 0.931 -> 0.956 | Improved |
| Add unpaired image data (ablation) | BMA-Memory weakly paired associative learning · ACIVW sound-to-image retrieval | Rank 1 retrieval accuracy: 0.498 -> 0.522 | Improved |
| Add unpaired image data from Kinetics-400 (ablation) | BMA-Memory weakly paired associative learning · DCASE-2018 zero-shot sound classification | Top-1 accuracy (linear evaluation): 0.538 -> 0.562 | Improved |
| Add unpaired sound while retaining only 20% of paired data (ablation) | BMA-Memory weakly paired associative learning · ACIVW image classification (20% paired data) | Top-1 accuracy: 0.693 -> 0.749 | Improved |
| Add unpaired sound to full paired-data training (ablation) | BMA-Memory weakly paired associative learning · ACIVW image classification (100% paired data) | Top-1 accuracy: 0.745 -> 0.778 | Improved |
| Increase unpaired sound volume (ablation) | BMA-Memory weakly paired associative learning · ACIVW image classification | Top-1 accuracy | Mixed |
| Increase BMA-Memory slot count (ablation) | BMA-Memory · ACIVW image classification | Top-1 accuracy | Improved |
Closest papers by meaning
- Multi-Modality Associative Bridging Through Memory: Speech Sound Recollected From Face Video (ICCV 2021 · similarity 0.88)
- Learning Compositional Sparse Bimodal Models (TPAMI 2017 · similarity 0.86)
- Weakly-supervised Audio Separation via Bi-modal Semantic Similarity (ICLR 2024 · similarity 0.84)
- Asymmetric Reinforcing Against Multi-Modal Representation Bias (AAAI 2025 · similarity 0.83)
- Gramian Multimodal Representation Learning and Alignment (ICLR 2025 · similarity 0.83)
- Aligning Multimodal Representations through an Information Bottleneck (ICML 2025 · similarity 0.83)
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.
