GLUE / SuperGLUE
← Most-used data · All Text and NLP benchmarks datasets
Corpora and benchmarks for language understanding, question answering and text classification.
Tasks, access and usage conditions#
Mention counts below measure name matches in tracked papers, not benchmark quality or downloads. A family count can cover multiple datasets; it is not a separate count for each comparison row. Choose a task and version first, preserve evaluation splits, and follow the linked release terms. The site's content license does not replace the datasets' own terms.
| Dataset / version | Task and evaluation | Access and preparation | Usage conditions |
|---|---|---|---|
| GLUE | A suite spanning sentiment, acceptability, paraphrase, similarity and inference. Evaluation: Report the prescribed score for each task and suite aggregate; STS-B, MRPC and other tasks use different metrics. | Official source / documentation Use official task downloads and splits; several test labels are hidden for benchmark evaluation. | Component datasets retain their original terms; there is no single reuse license implied by a GLUE score. |
| SuperGLUE | A separate suite emphasizing reading comprehension, reasoning and difficult language understanding. Evaluation: Use each task’s specified metrics and the official aggregate; it is not directly interchangeable with GLUE. | Official source / documentation Use official task-specific downloads; keep training, validation and submitted test evaluation distinct. | Review each component’s license and source-text conditions separately, following the benchmark FAQ and original releases. |
Key findings#
- From Nov 2023 to Oct 2026, 98 archived papers mentioned GLUE / SuperGLUE in their title, abstract or keywords (the search of earlier months' full texts is still under way), most of them in Dec 2024 (7).
- No paper of the tracked journals mentioned GLUE / SuperGLUE in the last 30 days (2026-10-06 to 2026-11-04).
- From 2017 to 2026 (so far), GLUE / SuperGLUE was mentioned in 108 papers in the tracked journals, most of them in AI (100), information systems (4) and operations research (4).
- 32 of them were published in 2025, against 0 in 2017.
- Since 2023, the institutions with the most of these papers were Chinese Academy of Sciences (10), Institute of Automation (6) and Czech Technical University in Prague (3).
- The journals publishing most of them were Neural Networks (12), IEEE Transactions on Pattern Analysis and Machine Intelligence (11) and Knowledge-Based Systems (9).
- The most-cited of them is “A Survey on Deep Neural Network Pruning: Taxonomy, Comparison, Analysis, and Recommendations” (IEEE Transactions on Pattern Analysis and Machine Intelligence, 2024), with 512 citations.
- Papers use GLUE / SuperGLUE most often with Hugging Face Hub (26), Natural language inference (SNLI, MultiNLI) (24) and Stanford Sentiment Treebank (19); it has 28 partners in the code and data network, in the group around Twitter / X, Facebook, YouTube / TikTok.
Mentions over 36 months#
| Trend | Nov 23–Oct 24 | Nov 24–Oct 25 | Nov 25–Oct 26 | 36 months |
|---|---|---|---|---|
| 32 | 33 | 33 | 98 |
Papers per year#
20172026 (so far)
| Category | 2017 | 2018 | 2019 | 2020 | 2021 | 2022 | 2023 | 2024 | 2025 | 2026 (so far) |
|---|---|---|---|---|---|---|---|---|---|---|
| AI | 0 | 0 | 0 | 1 | 1 | 1 | 12 | 33 | 29 | 23 |
| Information Systems | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 2 | 2 |
| Operations Research | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 2 | 1 | 1 |
| All categories | 0 | 0 | 0 | 1 | 1 | 1 | 12 | 35 | 32 | 26 |
Where the papers come from#
| # | Journal | Papers |
|---|---|---|
| 1 | Neural Networks | 12 |
| 2 | IEEE Transactions on Pattern Analysis and Machine Intelligence | 11 |
| 3 | Knowledge-Based Systems | 9 |
| 4 | International Journal of Computer Vision | 8 |
| 5 | Artificial Intelligence Review | 6 |
| 6 | Expert Systems with Applications | 6 |
| 7 | Language Resources and Evaluation | 5 |
| 8 | Pattern Recognition | 4 |
| 9 | Computer Speech & Language | 4 |
| 10 | Machine Learning | 4 |
| # | Institution | Papers |
|---|---|---|
| 1 | Chinese Academy of Sciences | 10 |
| 2 | Institute of Automation | 6 |
| 3 | Czech Technical University in Prague | 3 |
| 4 | Korea Advanced Institute of Science and Technology | 3 |
| 5 | Meta (United States) | 3 |
| 6 | National University of Defense Technology | 3 |
| 7 | Peking University | 3 |
| 8 | Shanghai Jiao Tong University | 3 |
| 9 | Technical University of Munich | 3 |
| 10 | Tencent (China) | 3 |
| # | Country | Papers |
|---|---|---|
| 1 | China | 47 |
| 2 | United States | 12 |
| 3 | Australia | 10 |
| 4 | United Kingdom | 10 |
| 5 | Italy | 9 |
| 6 | Germany | 8 |
| 7 | South Korea | 7 |
| 8 | Switzerland | 4 |
| 9 | Hong Kong | 4 |
| 10 | India | 4 |
| # | Keyword | Papers |
|---|---|---|
| 1 | large language models | 26 |
| 2 | BERT | 17 |
| 3 | GLUE benchmark | 14 |
| 4 | pre-trained language models | 11 |
| 5 | transformer models | 10 |
| 6 | parameter-efficient fine-tuning | 8 |
| 7 | self-supervised learning | 7 |
| 8 | low-rank adaptation | 6 |
| 9 | model compression | 6 |
| 10 | natural language processing | 6 |
Most-cited papers#
Used together with#
| Used with | Type | Papers using both | Lift |
|---|---|---|---|
| Hugging Face Hub | Dataset | 26 | 24.7 |
| Natural language inference (SNLI, MultiNLI) | Dataset | 24 | 113.1 |
| Stanford Sentiment Treebank | Dataset | 19 | 145.1 |
| Wikipedia | Dataset | 18 | 7.8 |
| ImageNet | Dataset | 16 | 3.0 |
| CIFAR-10 / CIFAR-100 | Dataset | 14 | 4.5 |
| Twitter / X | Dataset | 12 | 1.2 |
| CoNLL | Dataset | 11 | 38.4 |
| BIG-bench | Dataset | 8 | 114.3 |
| IMDb | Dataset | 8 | 15.1 |
| MMLU | Dataset | 8 | 44.9 |
| Pre-training corpora (Common Crawl, C4, The Pile) | Dataset | 8 | 32.2 |
Cite this page
Scholar4AI. (2026). GLUE / SuperGLUE. Journal and Conference Research Rankings. Retrieved November 4, 2026, from https://scholar4ai.com/datasets/glue-superglue
@misc{scholar4ai-datasets-glue-superglue,
author = {{Scholar4AI}},
title = {GLUE / SuperGLUE},
year = {2026},
howpublished = {\url{https://scholar4ai.com/datasets/glue-superglue}},
urldate = {2026-11-04},
note = {Journal and Conference Research Rankings; data from OpenAlex (CC0); licensed CC BY-NC 4.0}
}