Research map / Vision-language and multimodal / Cross-modal learning
Efficient Large-Scale Multi-Modal Classification
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 continuous visual features using multimodal fusion | FastText · Food101 | accuracy: 88.0 $\pm$ 0.1 -> 90.8 $\pm$ 0.1 | Helped |
| Add continuous visual features using multimodal fusion | FastText · MM-IMDB | accuracy: 58.8 $\pm$ 0.1 -> 62.3 $\pm$ 0.2 | Helped |
| Add continuous visual features using multimodal fusion | FastText · FlickrTag-1 | accuracy: 23.0 $\pm$ 0.0 -> 28.6 $\pm$ 0.0 | Helped |
| Use discretized visual features rather than continuous visual features | Multimodal classification · FlickrTag-1 | train time: 1h06m -> 0h02m | Helped |
| Add product-quantized visual features | FastText · FlickrTag | accuracy: 36.7 -> 38.9 | Helped |
| Use random sample product quantization instead of product quantization | Discretized multimodal classifier · FlickrTag | accuracy: 38.9 -> 39.4 | Helped |
| Use bilinear-gated fusion instead of additive fusion | Continuous multimodal classifier · Food101 | accuracy: 90.4 $\pm$ 0.1 -> 90.8 $\pm$ 0.1 | Helped |
| Use bilinear-gated fusion instead of additive fusion | Continuous multimodal classifier · MM-IMDB | accuracy: 61.0 $\pm$ 0.0 -> 62.3 $\pm$ 0.2 | Helped |
| Use bilinear-gated fusion instead of additive fusion | Continuous multimodal classifier · FlickrTag-1 | accuracy: 26.8 $\pm$ 0.0 -> 28.6 $\pm$ 0.0 | Helped |
| Replace continuous visual feature processing with product-quantized tokens | Multimodal classifier · FlickrTag-1 | train time: 1h04m -> 0h01m | Helped |
Closest papers by meaning
- Beyond Modality Fusion: Deep Ensembles for Multimodal Classification (NeurIPS 2026 · similarity 0.88)
- Semi-Supervised Multimodal Classification Through Learning from Modal and Strategic Complementarities (AAAI 2025 · similarity 0.87)
- On the Comparison between Multi-modal and Single-modal Contrastive Learning (NeurIPS 2024 · similarity 0.86)
- Facilitating Multimodal Classification via Dynamically Learning Modality Gap (NeurIPS 2024 · similarity 0.85)
- Amplifying Prominent Representations in Multimodal Learning via Variational Dirichlet Process (NeurIPS 2025 · similarity 0.85)
- Multi-Layer Visual Feature Fusion in Multimodal LLMs: Methods, Analysis, and Best Practices (CVPR 2025 · similarity 0.85)
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.