Research map / Vision-language and multimodal / Cross-modal learning
MultiBench: Multiscale Benchmarks for Multimodal Representation Learning
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 |
|---|---|---|---|
| apply out-of-domain methods instead of in-domain | MultiZoo methods · MUStARD | Acc: 66.3 -> 71.8 | Helped |
| apply out-of-domain methods | MultiZoo methods · UR-FUNNY | Acc: 62.9 -> 66.7 | Helped |
| apply out-of-domain methods | MultiZoo methods · V&T EE | MSE: 0.258 -> 0.185 | Helped |
| apply out-of-domain methods | MultiZoo methods · ENRICO | Acc: 47.0 -> 51.0 | Helped |
| apply out-of-domain methods | MultiZoo methods · CMU-MOSI | Acc: 83.0 -> 75.5 | Hurt |
| apply out-of-domain methods | MultiZoo methods · Kinetics-S | Acc: 56.1 -> 23.7 | Hurt |
| apply out-of-domain methods | MultiZoo methods · MuJoCo Push | MSE: 0.290 -> 0.402 | Hurt |
| multi-dataset multitask training across 4 affective datasets, late fusion | Late fusion · CMU-MOSEI | Acc: 79.2 -> 80.9 | Helped |
| multi-dataset multitask training across 4 affective datasets | Late fusion · CMU-MOSI | Acc: 75.2 -> 70.8 | Hurt |
| multi-dataset multitask training, multimodal transformer | MulT · CMU-MOSEI | Acc: 82.1 -> 82.9 | Helped |
| multimodal vs best unimodal | multimodal models · MultiBench | parameters: 2-3x parameters | Mixed |
Closest papers by meaning
- Fusion or Confusion? Multimodal Complexity Is Not All You Need (NeurIPS 2026 · similarity 0.90)
- Amplifying Prominent Representations in Multimodal Learning via Variational Dirichlet Process (NeurIPS 2025 · similarity 0.89)
- MEGA-Bench: Scaling Multimodal Evaluation to over 500 Real-World Tasks (ICLR 2025 · similarity 0.89)
- Asymmetric Reinforcing Against Multi-Modal Representation Bias (AAAI 2025 · similarity 0.89)
- Continual Multimodal Contrastive Learning (NeurIPS 2025 · similarity 0.89)
- Multimodal Representation Learning by Alternating Unimodal Adaptation (CVPR 2024 · similarity 0.88)
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.