AAAI 2024 accepted papers by research area

2,864 papers were accepted at AAAI 2024 (+841 vs 2023). Here is how they split across 20 research areas, and which topics gained the most ground.

Research areas

Area	Papers	Share	Change in share vs 2023
Label-efficient and robust learning	399	13.9%	-3.1 pts
Learning theory and algorithms	291	10.2%	-1.2 pts
Large language models	239	8.3%	+2.2 pts
Image generation	218	7.6%	+3.4 pts
Graph learning	201	7.0%	-1.5 pts
Vision-language and multimodal	198	6.9%	+2.6 pts
Probabilistic and scientific modeling	160	5.6%	-0.8 pts
Detection and segmentation	152	5.3%	+0.4 pts
Architectures and efficiency	152	5.3%	-1.0 pts
Reinforcement learning	137	4.8%	-0.9 pts
3D vision	114	4.0%	+1.6 pts
Video	112	3.9%	steady
Multi-agent systems and games	93	3.2%	-1.8 pts
Adversarial robustness and security	93	3.2%	-0.3 pts
Federated learning and privacy	90	3.1%	+1.0 pts
Point clouds	54	1.9%	+0.4 pts
Continual learning	49	1.7%	+0.2 pts
Time series	46	1.6%	-0.2 pts
Bandits and online learning	33	1.2%	-0.6 pts
Optimization	33	1.2%	-0.2 pts

Change in share of AAAI papers from 2023 to 2024, topics with at least 10 papers.

Language, safety and interpretability (Large language models): 32 → 86 papers
Text-to-image diffusion (Image generation): 0 → 36 papers
Personalized and heterogeneous FL (Federated learning and privacy): 11 → 42 papers
Human and embodied modeling (Vision-language and multimodal): 39 → 78 papers
CLIP and zero-shot (Vision-language and multimodal): 13 → 41 papers
Fast diffusion sampling (Image generation): 3 → 25 papers
3D generation (3D vision): 11 → 35 papers
Editing and style transfer (Image generation): 18 → 43 papers
Cross-modal learning (Vision-language and multimodal): 23 → 48 papers
NeRF and radiance fields (3D vision): 5 → 22 papers

AAAI in other years: 2023 · 2025 · 2026