AI research rankings
Papers in the 24 tracked AI journals, window 2026-09-27 – 2026-10-26 (30 days). Keywords, institutions, web domains and countries are ranked by paper count. Also as Markdown.
1,124Papers (30 days)
80Papers (last 7 days)
24Journals
1,176Institutions
74Countries
1,103Web domains
Key findings
- In the last 30 days (2026-09-27 to 2026-10-26), 1,124 papers were published in the 24 tracked AI journals (as indexed by OpenAlex).
- The most frequent keywords were “large language models” (70), “graph neural networks” (38) and “multimodal fusion” (29).
- The institutions with the most papers were Chinese Academy of Sciences (30), Tsinghua University (25) and University of Electronic Science and Technology of China (22).
- China was involved in 63% of the papers (712), followed by United States (55) and India (50).
- Among institution web domains, .edu.cn appeared on 61% of the papers.
- The most-mentioned equations and models were Graph neural network (graph convolution) (73), Scaled dot-product attention (Transformer) (55) and Contrastive loss (InfoNCE) (39).
- The most-mentioned kinds of data were Image classification benchmarks (8), Object detection and segmentation benchmarks (3) and Social media data (3); the most-mentioned datasets were MNIST / Fashion-MNIST (5), CIFAR-10 / CIFAR-100 (4) and ImageNet (3).
Papers per day
2026-09-272026-10-26
Show the numbers
| Date | Papers |
|---|---|
| 2026-09-27 | 15 |
| 2026-09-28 | 36 |
| 2026-09-29 | 42 |
| 2026-09-30 | 56 |
| 2026-10-01 | 366 |
| 2026-10-02 | 14 |
| 2026-10-03 | 35 |
| 2026-10-04 | 4 |
| 2026-10-05 | 21 |
| 2026-10-06 | 34 |
| 2026-10-07 | 40 |
| 2026-10-08 | 34 |
| 2026-10-09 | 44 |
| 2026-10-10 | 25 |
| 2026-10-11 | 13 |
| 2026-10-12 | 29 |
| 2026-10-13 | 36 |
| 2026-10-14 | 50 |
| 2026-10-15 | 41 |
| 2026-10-16 | 24 |
| 2026-10-17 | 46 |
| 2026-10-18 | 10 |
| 2026-10-19 | 29 |
| 2026-10-20 | 22 |
| 2026-10-21 | 21 |
| 2026-10-22 | 13 |
| 2026-10-23 | 20 |
| 2026-10-24 | 4 |
| 2026-10-25 | 0 |
| 2026-10-26 | 0 |
Growing over the year Trends →
| Keyword | Papers per month, Oct 25 – Oct 26* | Papers Jul 2026–Sep 2026 | vs. earlier share |
|---|---|---|---|
| AI governance | 41 | +18 | |
| mixture of experts | 47 | +17 | |
| uncertainty estimation | 39 | +16 | |
| foundation models | 27 | +11 | |
| spatial transcriptomics | 14 | +11 |
Top research keywords
| # | Keyword | Papers | Last 7 days |
|---|---|---|---|
| 1 | large language models | 70 | 4 |
| 2 | graph neural networks | 38 | 1 |
| 3 | multimodal fusion | 29 | 0 |
| 4 | contrastive learning | 27 | 0 |
| 5 | explainable AI | 24 | 2 |
| 6 | knowledge distillation | 23 | 1 |
| 7 | federated learning | 21 | 4 |
| 8 | class imbalance | 19 | 1 |
| 8 | vision-language models | 19 | 0 |
| 10 | diffusion models | 18 | 1 |
| 10 | feature selection | 18 | 1 |
| 10 | transformer models | 18 | 2 |
| 13 | interpretability | 15 | 1 |
| 13 | reinforcement learning | 15 | 0 |
| 15 | attention mechanism | 14 | 2 |
| 15 | deep learning | 14 | 2 |
| 15 | domain adaptation | 14 | 1 |
| 15 | retrieval-augmented generation | 14 | 1 |
| 19 | convolutional neural networks | 13 | 1 |
| 20 | computational efficiency | 12 | 1 |
Top universities and research institutions
| # | Institution | Papers | Last 7 days |
|---|---|---|---|
| 1 | Chinese Academy of Sciences | 30 | 5 |
| 2 | Tsinghua University | 25 | 4 |
| 3 | University of Electronic Science and Technology of China | 22 | 1 |
| 3 | National University of Defense Technology | 22 | 0 |
| 5 | Beijing Institute of Technology | 21 | 1 |
| 5 | Wuhan University | 21 | 0 |
| 7 | University of Chinese Academy of Sciences | 20 | 3 |
| 8 | Harbin Institute of Technology | 19 | 2 |
| 9 | Beijing Jiaotong University | 17 | 1 |
| 10 | Southeast University | 16 | 1 |
| 10 | Northeastern University | 16 | 2 |
| 12 | Xidian University | 15 | 0 |
| 13 | Sun Yat-sen University | 13 | 1 |
| 13 | Shanghai Jiao Tong University | 13 | 3 |
| 13 | Xi'an Jiaotong University | 13 | 1 |
| 16 | Central South University | 12 | 0 |
| 16 | Hong Kong Polytechnic University | 12 | 0 |
| 16 | Anhui University | 12 | 0 |
| 16 | Northwestern Polytechnical University | 12 | 0 |
| 16 | Fudan University | 12 | 2 |
Top institution web domains
| # | Web domain | Papers | Last 7 days |
|---|---|---|---|
| 1 | cas.cn | 37 | 5 |
| 2 | tsinghua.edu.cn | 26 | 4 |
| 3 | nudt.edu.cn | 22 | 0 |
| 3 | uestc.edu.cn | 22 | 1 |
| 3 | whu.edu.cn | 22 | 0 |
| 6 | bit.edu.cn | 21 | 1 |
| 7 | ucas.ac.cn | 20 | 3 |
| 8 | hit.edu.cn | 19 | 2 |
| 9 | njtu.edu.cn | 17 | 1 |
| 9 | seu.edu.cn | 17 | 1 |
| 11 | neu.edu.cn | 16 | 2 |
| 12 | xidian.edu.cn | 15 | 0 |
| 13 | sjtu.edu.cn | 13 | 3 |
| 13 | sysu.edu.cn | 13 | 1 |
| 13 | xjtu.edu.cn | 13 | 1 |
| 16 | ahu.edu.cn | 12 | 0 |
| 16 | csu.edu.cn | 12 | 0 |
| 16 | fudan.edu.cn | 12 | 2 |
| 16 | nwpu.edu.cn | 12 | 0 |
| 16 | polyu.edu.hk | 12 | 0 |
Top countries
| # | Country | Papers | Last 7 days |
|---|---|---|---|
| 1 | China (CN) | 712 | 49 |
| 2 | United States (US) | 55 | 7 |
| 3 | India (IN) | 50 | 2 |
| 4 | United Kingdom (GB) | 43 | 4 |
| 5 | Hong Kong (HK) | 34 | 3 |
| 6 | Australia (AU) | 33 | 3 |
| 7 | South Korea (KR) | 29 | 1 |
| 8 | Singapore (SG) | 24 | 3 |
| 9 | Turkey (TR) | 22 | 0 |
| 10 | United Arab Emirates (AE) | 21 | 4 |
| 11 | Italy (IT) | 19 | 0 |
| 12 | Canada (CA) | 15 | 2 |
| 12 | Iran (IR) | 15 | 0 |
| 14 | Spain (ES) | 14 | 0 |
| 14 | Malaysia (MY) | 14 | 0 |
| 14 | Saudi Arabia (SA) | 14 | 1 |
| 17 | Germany (DE) | 13 | 0 |
| 17 | Japan (JP) | 13 | 2 |
| 19 | Bangladesh (BD) | 11 | 1 |
| 19 | Egypt (EG) | 11 | 1 |
Research fields
| # | Field | Papers | Last 7 days |
|---|---|---|---|
| 1 | Computer Science | 690 | 43 |
| 2 | Engineering | 184 | 12 |
| 3 | Social Sciences | 34 | 1 |
| 4 | Medicine | 31 | 3 |
| 5 | Decision Sciences | 27 | 6 |
| 6 | Physics and Astronomy | 23 | 4 |
| 7 | Neuroscience | 19 | 1 |
| 8 | Psychology | 18 | 1 |
| 9 | Biochemistry, Genetics and Molecular Biology | 17 | 1 |
| 10 | Business, Management and Accounting | 15 | 2 |
| 11 | Agricultural and Biological Sciences | 8 | 0 |
| 12 | Mathematics | 7 | 1 |
| 13 | Materials Science | 5 | 1 |
| 14 | Arts and Humanities | 3 | 0 |
| 14 | Environmental Science | 3 | 0 |
| 14 | Health Professions | 3 | 0 |
| 17 | Economics, Econometrics and Finance | 2 | 0 |
| 18 | Dentistry | 1 | 0 |
| 18 | Earth and Planetary Sciences | 1 | 0 |
Papers per journal
| # | Journal | Papers | Last 7 days |
|---|---|---|---|
| 1 | Expert Systems with Applications | 249 | 14 |
| 2 | Pattern Recognition | 188 | 9 |
| 3 | Knowledge-Based Systems | 110 | 6 |
| 4 | Neural Computing and Applications | 90 | 0 |
| 5 | Neural Networks | 89 | 6 |
| 6 | Information Fusion | 71 | 1 |
| 7 | Applied Intelligence | 64 | 8 |
| 8 | Information Processing & Management | 54 | 2 |
| 9 | International Journal of Computer Vision | 39 | 10 |
| 10 | Artificial Intelligence Review | 28 | 9 |
| 10 | Knowledge and Information Systems | 28 | 4 |
| 12 | Machine Learning | 20 | 1 |
| 13 | Journal Of Big Data | 18 | 1 |
| 14 | Nature Machine Intelligence | 17 | 5 |
| 15 | AI & Society | 11 | 0 |
| 15 | Artificial Intelligence | 11 | 1 |
| 17 | Cognitive Computation | 10 | 3 |
| 18 | Data Mining and Knowledge Discovery | 8 | 0 |
| 19 | IEEE Transactions on Neural Networks and Learning Systems | 5 | 0 |
| 20 | IEEE Transactions on Evolutionary Computation | 4 | 0 |
Most-used equations and models
| # | Equation or model | Formula | Papers |
|---|---|---|---|
| 1 | Graph neural network (graph convolution) | 73 | |
| 2 | Scaled dot-product attention (Transformer) | 55 | |
| 3 | Contrastive loss (InfoNCE) | 39 | |
| 4 | Convolutional neural network (CNN) | 30 | |
| 5 | Logistic regression (logit) | 25 | |
| 6 | Long short-term memory (LSTM) | 15 | |
| 7 | Gradient boosting (XGBoost) | 10 | |
| 7 | Shapley value (SHAP) | 10 | |
| 9 | Random forest | 9 | |
| 9 | Support vector machine (SVM) | 9 |
Most-used data
| # | Data category | Most-mentioned datasets | Papers |
|---|---|---|---|
| 1 | Image classification benchmarks | MNIST / Fashion-MNIST (5), CIFAR-10 / CIFAR-100 (4), ImageNet (3) | 8 |
| 2 | Object detection and segmentation benchmarks | KITTI (3), nuScenes / Waymo Open / Argoverse (1) | 3 |
| 2 | Social media data | Twitter / X (2), Facebook (2), Reddit (1) | 3 |
| 4 | China household and social surveys | China Household Finance Survey (CHFS) (1), China Labor-force Dynamics Survey (CLDS) (1) | 2 |
| 4 | Credit scoring and lending data | German Credit (UCI Statlog) (2) | 2 |
| 4 | Face, person and video benchmarks | Face recognition (LFW, CelebA, VGGFace2) (1), Action recognition (Kinetics, UCF101, HMDB51) (1) | 2 |
| 4 | Medical and clinical datasets | MIMIC (1), ISIC skin lesion archive (1) | 2 |
| 4 | Mutual fund and hedge fund data | Lipper / TASS (2) | 2 |
| 4 | Online reviews, e-commerce and recommender data | Amazon reviews (1), Taobao / Tmall / JD.com (1) | 2 |
| 4 | Reinforcement learning environments | D4RL (1), SUMO traffic simulator (1) | 2 |
Latest AI papers
Questions this page answers
- What are the most common topics in recent AI research?
- In the last 30 days, the most frequent OpenAlex keywords in papers from the tracked AI journals were “large language models” (70), “graph neural networks” (38), “multimodal fusion” (29), “contrastive learning” (27) and “explainable AI” (24).
- Which universities and institutions publish the most AI research?
- In the last 30 days, the institutions with the most papers in the tracked AI journals were Chinese Academy of Sciences (30), Tsinghua University (25), University of Electronic Science and Technology of China (22), National University of Defense Technology (22) and Beijing Institute of Technology (21).
- Which countries publish the most AI research?
- Counting papers with at least one author at an institution in the country, the leaders over the last 30 days were China (712), United States (55), India (50), United Kingdom (43) and Hong Kong (34).
- Which equations and models do recent AI research papers use most?
- Counting papers that mention them by name in the title, abstract, keywords or full text, the most frequent in the tracked AI journals over the last 30 days were Graph neural network (graph convolution) (73), Scaled dot-product attention (Transformer) (55), Contrastive loss (InfoNCE) (39), Convolutional neural network (CNN) (30) and Logistic regression (logit) (25).
- What data do recent AI research papers use?
- Counting papers that mention a dataset by name, the kinds of data used most over the last 30 days were Image classification benchmarks (8), Object detection and segmentation benchmarks (3), Social media data (3), China household and social surveys (2) and Credit scoring and lending data (2); the most-mentioned datasets were MNIST / Fashion-MNIST (5), CIFAR-10 / CIFAR-100 (4), ImageNet (3), KITTI (3) and Facebook (2).
- How current is this data?
- It is rebuilt from OpenAlex 4 times a day (00:00, 06:00, 12:00, 18:00, UTC+09:00); the last update was 2026-10-26 11:39 UTC+09:00. OpenAlex usually indexes new papers within a few days of publication.