ICLR 2024 accepted papers by research area
2,259 papers were accepted at ICLR 2024 (+684 vs 2023). Here is how they split across 20 research areas, and which topics gained the most ground.
See ICLR 2024 in the interactive map
Research areas
| Area | Papers | Share | Change in share vs 2023 |
|---|---|---|---|
| Large language models | 328 | 14.5% | +9.3 pts |
| Label-efficient and robust learning | 221 | 9.8% | -6.0 pts |
| Architectures and efficiency | 221 | 9.8% | -0.8 pts |
| Probabilistic and scientific modeling | 196 | 8.7% | -0.7 pts |
| Reinforcement learning | 187 | 8.3% | -1.1 pts |
| Image generation | 185 | 8.2% | +3.4 pts |
| Learning theory and algorithms | 130 | 5.8% | -2.1 pts |
| Graph learning | 123 | 5.4% | -1.5 pts |
| Vision-language and multimodal | 116 | 5.1% | +1.8 pts |
| Adversarial robustness and security | 81 | 3.6% | steady |
| 3D vision | 74 | 3.3% | +0.4 pts |
| Optimization | 72 | 3.2% | -1.6 pts |
| Federated learning and privacy | 62 | 2.7% | -0.4 pts |
| Multi-agent systems and games | 53 | 2.3% | steady |
| Time series | 52 | 2.3% | +0.8 pts |
| Video | 46 | 2.0% | +0.6 pts |
| Detection and segmentation | 42 | 1.9% | -1.5 pts |
| Continual learning | 27 | 1.2% | -0.5 pts |
| Bandits and online learning | 25 | 1.1% | -0.2 pts |
| Point clouds | 18 | 0.8% | steady |
Fastest-growing topics at ICLR 2024
Change in share of ICLR papers from 2023 to 2024, topics with at least 10 papers.
- Language, safety and interpretability (Large language models): 46 → 121 papers
- Instruction and fine-tuning (Large language models): 1 → 53 papers
- Agents, code and math (Large language models): 16 → 65 papers
- Autoregressive and latent generation (Image generation): 18 → 56 papers
- Text-to-image diffusion (Image generation): 7 → 39 papers
- Retrieval and knowledge (Large language models): 9 → 38 papers
- CLIP and zero-shot (Vision-language and multimodal): 15 → 45 papers
- Alignment and preferences (Large language models): 2 → 24 papers
- Fast diffusion sampling (Image generation): 27 → 58 papers
- 3D generation (3D vision): 9 → 31 papers
More venues
Same year: ICML 2024 · NeurIPS 2024 · CVPR 2024 · AAAI 2024