You Only Look Once: Unified, Real-Time Object Detection
Open accessThe DOI matches its Crossref record; 53 of the 53 references it cites were found in a registry or an index; 8 of the 25 sources its claims cite were checked for support.
Abstract
We present YOLO, a new approach to object detection. Prior work on object detection repurposes classifiers to perform detection. Instead, we frame object detection as a regression problem to spatially separated bounding boxes and associated class probabilities. A single neural network predicts bounding boxes and class probabilities directly from full images in one evaluation. Since the whole detection pipeline is a single network, it can be optimized end-to-end directly on detection performance. Our unified architecture is extremely fast. Our base YOLO model processes images in real-time at 45 frames per second. A smaller version of the network, Fast YOLO, processes an astounding 155 frames per second while still achieving double the mAP of other real-time detectors. Compared to state-of-the-art detection systems, YOLO makes more localization errors but is less likely to predict false positives on background. Finally, YOLO learns very general representations of objects. It outperforms other detection methods, including DPM and R-CNN, when generalizing from natural images to other domains like artwork.
Overview
CoChat Summary
TL;DR
YOLO reframes object detection as a single regression problem, using one convolutional network to predict bounding boxes and class probabilities directly from full images. It achieves real-time detection speeds while remaining competitive with, and complementary to, slower region-based detectors.
Key findings
- YOLO runs at 45 FPS on a Titan X GPU, and Fast YOLO runs at 155 FPS; Fast YOLO reaches 52.7% mAP on PASCAL VOC 2007, more than double prior real-time detectors, while YOLO reaches 63.4% mAP.
- On PASCAL VOC 2012, YOLO scores 57.9% mAP; combining Fast R-CNN with YOLO raises Fast R-CNN from 71.8% to 75.0% mAP on VOC 2007, a 3.2% gain.
- YOLO makes fewer background false positives than Fast R-CNN: Fast R-CNN has 13.6% background false positives among top detections and is almost 3× more likely to predict background detections than YOLO.
- YOLO generalizes better to artwork: on Picasso, YOLO gets 53.3 AP versus R-CNN’s 10.4 AP and DPM’s 37.8 AP; on People-Art, YOLO gets 45 AP versus R-CNN’s 26 AP and DPM’s 32 AP.
Methods
- Detection is modeled as a single CNN regression: the image is divided into an S×S grid; for PASCAL VOC, S=7, B=2, C=20, producing a 7×7×30 tensor of bounding boxes, confidence scores, and class probabilities.
- The base network has 24 convolutional layers followed by 2 fully connected layers; Fast YOLO uses 9 convolutional layers. Convolutional layers are pretrained on ImageNet at 224×224, then detection training uses 448×448 input.
- Training uses PASCAL VOC 2007 and 2012 for about 135 epochs, batch size 64, momentum 0.9, decay 0.0005, sum-squared error with λ_coord=5 and λ_noobj=0.5, dropout, and data augmentation.
Limitations
- The paper states YOLO imposes strong spatial constraints because each grid cell predicts only two boxes and one class, limiting nearby objects and struggling with small objects in groups such as flocks of birds.
- The paper states YOLO struggles to generalize to objects in new or unusual aspect ratios or configurations, uses relatively coarse features due to downsampling, and its loss treats small and large box errors equally; its main error source is incorrect localization.
- The paper states YOLO still lags state-of-the-art detection accuracy and struggles to precisely localize some objects, especially small ones.
Model-written summary. Check it against the paper before relying on it.
Integrity signals
Registry
Cites
Statistics
Figures





Verification
Claims the paper makes
6 Valid · 8 Not valid · 15 Not checked · 92 Own results and interpretations
The cited sources that could be checked couldn’t confirm or dispute this claim.
In the paper
“While it is much faster than R-CNN, it still falls short of real-time and takes a significant accuracy hit from not having good proposals.”
What the sources say
- Could not verifycites [20] [20] K. Lenc and A. Vedaldi. R-cnn minus r. arXiv preprint arXiv:1506.06981 , 2015. 5, 6 - sourceExperimental — accuracy not yet measured
The ConvNet-based Object Detection section identifies R-CNN minus R as a method that uses predefined regions instead of region proposals. It gives no runtime measurement or assessment of whether the method achieves real-time detection; searches for real-time performance and speed did not establish that claim. (LLM, conf=0.76)
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Could not verify
The cited sources support only part of this claim — read each source’s verdict below.
In the paper
“To avoid overfitting we use dropout and extensive data augmentation.”
What the sources say
- Partly supportedcites [18] [18] G. E. Hinton, N. Srivastava, A. Krizhevsky, I. Sutskever, and R. R. Salakhutdinov. Improving neural networks by preventing co-adaptation of feature detectors. arXiv preprint arXiv:1207.0580 , 2012. 4 - sourceExperimental — accuracy not yet measured
The paper states that dropout reduces overfitting (p. 1) and describes using dropout with data augmentation for its ImageNet model (H Models for ImageNet, p. 18). It does not establish that YOLO uses either technique or describe YOLO’s augmentation as extensive. (LLM, conf=0.88)
plainstates a cause- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Partly supported
Every source the paper cites for this claim was checked and supports it.
In the paper
“However, only 30Hz DPM [31] actually runs in real-time.”
What the sources say
- Supportedcites [31] [31] M. A. Sadeghi and D. Forsyth. 30hz object detection with dpm v5. In Computer Vision-ECCV 2014 , pages 65-79. Springer, 2014. 5, 6 - source
The Introduction (p.1) states that 'except for certain objects including face, pedestrians and certain rigid objects, detectors do not currently run at video rate' and that the presented architecture detects the 20 PASCAL VOC categories simultaneously at 30Hz, i.e. at video/real-time rate. The Abstract (p.1) and Table 2 (Method) bound the comparison to other DPM speed-up efforts: reference DPM at 0.07Hz, the fastest GPU 'Vanilla DPM' at ~1Hz, FTVQ at 0.53s, and 'over an order of magnitude faster than the previous fastest DPM implementation', so none of the listed variants reaches real-time. Thus our copy supports both parts of the attribution: that this 30Hz DPM runs in real-time and that it is the only such DPM among the speed-up efforts compared. (LLM, conf=0.82)
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Supported
The cited sources that could be checked couldn’t confirm or dispute this claim.
In the paper
“While it is much faster than R-CNN, it still falls short of real-time and takes a significant accuracy hit from not having good proposals.”
What the sources say
- Could not verifycites [20] [20] K. Lenc and A. Vedaldi. R-cnn minus r. arXiv preprint arXiv:1506.06981 , 2015. 5, 6 - sourceExperimental — accuracy not yet measured
The ConvNet-based Object Detection section identifies R-CNN minus R and says it uses predefined regions in an SPPnet system, but gives no speed comparison with R-CNN. A search for R-CNN minus R and speed found no passage in our copy establishing that it is much faster. (LLM, conf=0.73)
plainstates a cause- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Could not verify
The cited sources support only part of this claim — read each source’s verdict below.
In the paper
“While it is much faster than R-CNN, it still falls short of real-time and takes a significant accuracy hit from not having good proposals.”
What the sources say
- Partly supportedcites [20] [20] K. Lenc and A. Vedaldi. R-cnn minus r. arXiv preprint arXiv:1506.06981 , 2015. 5, 6 - sourceExperimental — accuracy not yet measured
The “ConvNet-based Object Detection” section (page unavailable) says R-CNN minus R uses predefined regions instead of region proposals. It does not say that this causes a significant accuracy loss; the paper’s discussion of low accuracy concerns a different system, naïve Faster R-CNN. (LLM, conf=0.83)
plainstates a cause- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Partly supported
84 more claims
Create a free account to see all 121 claims and to verify the 2 references still open.
- Probably not a substantial claim. 17
- It may state two separate claims as one. 15
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
The source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“which was not thoroughly investigated in prevIoUs work such as YOLOv1 [20] and YOLOv2 [51] employing MSE/L2 loss (Mean Squared Error).”
From Object Detection in 20 Years: A Survey (2023) · cites this paper as [20]
Experimental — accuracy not yet measuredThe source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“the YOLO series models v2-v5 [144-147] are some of the most advanced object detection methods based on DNN, which have good performance in terms of speed and precision.”
From A survey of model compression strategies for object detection (2023) · cites this paper as [144]
Experimental — accuracy not yet measuredThe source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“As a typical and more widely used new form of applications [9], various deep learning-based intelligent services and applications have changed many aspects of people's lives due to the great advantages of Deep Learning (DL) in the fields of Computer Vision (CV) and Natural Language Processing (NLP) [10].”
From Convergence of Edge Computing and Deep Learning: A Comprehensive Survey (2020) · cites this paper as [9]
Experimental — accuracy not yet measuredThe cited source says what the claim attributes to it.
In the citing paper
“A convolutional neural network predicts the bounding boxes as well as class probabilities for all the objects depicted in an image.”
From Object detection using YOLO: challenges, architectural successors, datasets and applications (2023) · cites this paper as [56]
The cited source says what the claim attributes to it.
In the citing paper
“Authors of YOLO [56] have reframed the problem of object detection as a regression problem instead of classification problem.”
From Object detection using YOLO: challenges, architectural successors, datasets and applications (2023) · cites this paper as [56]
6 more claims about this paper
Create a free account to see every claim other papers make about this one.
Citations
Cites
Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks
Ren, S., He, K., Girshick, R., & Sun, J. (2017) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2016.2577031 · Cited by 73,472
Self-citationFaster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks
Ren, S., He, K., Girshick, R., & Sun, J. (2015) · arXiv · doi:10.48550/arxiv.1506.01497 · Cited by 70,812
Self-citationGoing deeper with convolutions
Szegedy, C., Liu, W., Jia, Y., Sermanet, P., Reed, S., Anguelov, D., Erhan, D., Vanhoucke, V., & Rabinovich, A. (2015) · 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2015.7298594 · Cited by 48,562
ImageNet Large Scale Visual Recognition Challenge
Russakovsky, O., Deng, J., Su, H., Krause, J., Satheesh, S., Ma, S., Huang, Z., Karpathy, A., Khosla, A., Bernstein, M., et al. (2015) · International Journal of Computer Vision · doi:10.1007/s11263-015-0816-y · Cited by 39,380 · cited 1×
“We pretrain our convolutional layers on the ImageNet 1000-class competition dataset [29].”
MethodHistograms of Oriented Gradients for Human Detection
Dalal, N., & Triggs, B. (2005) · 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'05) · doi:10.1109/cvpr.2005.177 · Cited by 33,584 · cited 2×
“Detection pipelines generally start by extracting a set of robust features from input images (Haar [25], SIFT [23], HOG [4], convolutional features [6]).”
MethodRich Feature Hierarchies for Accurate Object Detection and Semantic Segmentation
Girshick, R., Donahue, J., Darrell, T., & Malik, J. (2014) · 2014 IEEE Conference on Computer Vision and Pattern Recognition · doi:10.1109/cvpr.2014.81 · Cited by 32,684 · cited 2×
“Then, classifiers [35, 21, 13, 10] or localizers [1, 31] are used to identify objects in the feature space.”
MethodSelf-citationGirshick, R. (2015) · 2015 IEEE International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2015.169 · Cited by 28,105
Self-citationDelving Deep into Rectifiers: Surpassing Human-Level Performance on ImageNet Classification
He, K., Zhang, X., Ren, S., & Sun, J. (2015) · 2015 IEEE International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2015.123 · Cited by 19,741 · cited 1×
“Convolutional neural networks (CNNs) achieve impressive performance on classification tasks at real-time speeds [13].”
Object recognition from local scale-invariant features
Lowe, D. G. (1999) · Proceedings of the Seventh IEEE International Conference on Computer Vision · doi:10.1109/iccv.1999.790410 · Cited by 16,472 · cited 1×
“Detection pipelines generally start by extracting a set of robust features from input images (Haar [25], SIFT [23], HOG [4], convolutional features [6]).”
MethodRobust Real-Time Face Detection
Viola, P., & Jones, M. J. (2004) · International Journal of Computer Vision · doi:10.1023/b:visi.0000013087.49260.fb · Cited by 14,097 · cited 1×
“Detectors for single classes like faces or people can be highly optimized since they have to deal with much less variation [36].”
Background
Cited by
Items 46,101–46,200 of 51,399
Wernette, P. A., Esselman, P., Geglio, A., et al. (2025) · doi:10.23919/oceans59106.2025.11245182
“The geologic substrate model (i.e., RockFinder)[14] was based on a “you only look once” (YOLO)[15] framework, while the mussels model (i.e., MusselFinder)[16] was based on a SegFormer[17] framework.”
Li, K., Wan, Y., Yao, H., et al. (2025) · IEEE International Geoscience and Remote Sensing Symposium (IGARSS) · doi:10.1109/igarss55030.2025.11243366
“Traditional ANN-based object detection algorithms, such as Tiny YOLO[3], although effective in daytime scenarios, struggle to meet the requirements of nighttime missions.”
Xu, Y., Zhang, S., & Jiang, L. (2025) · IEEE International Geoscience and Remote Sensing Symposium (IGARSS) · doi:10.1109/igarss55030.2025.11243188
Sparse Attention Superyolo for Small Object Detection in Remote Sensing Imagery
Wang, Z., Chen, Y., & Wang, F. (2025) · IEEE International Geoscience and Remote Sensing Symposium (IGARSS) · doi:10.1109/igarss55030.2025.11243243
SonarCloud: A Simulated Forward Looking Sonar Dataset for Underwater Perception Tasks
Jaber, N., Al-Ramahi, Q., Christensen, L., & Wehbe, B. (2025) · doi:10.23919/oceans59106.2025.11245033
“The You Only Look Once (YOLO) series [51] , for instance, works by splitting the image into S × S grid of cells with each grid cell responsible for detecting an object if the center of that object falls within it.”
The Monocular Camera Measures the Edge Length of a Hexagonal Screw
Liu, M., Su, Y., Tan, S., et al. (2025) · doi:10.1109/cir65373.2025.11257412
“Although the YOLO series models are lightweight and real-time positioning[7][14], they only output bounding boxes and lack sub-pixel corner point information, which cannot meet the requirements of 0.01mm measurement[11].”
Kim, S., Park, J., Na, W., et al. (2025) · International Journal of Highway Engineering · doi:10.7855/ijhe.2025.27.5.119
Kapusi, T. P., Erdei, T. I., Abdullah, M., Husi, G., & Hajdú, A. (2025) · Robotics · doi:10.3390/robotics14120175
“Typical examples include the YOLO family, SSD (Single Shot Detector), and RetinaNet-based solutions [34–36].”
Object localization in images based on clustering of attention zones
Kosarevych, R., Lutsyk, O., Rusyn, B., et al. (2025) · Vìdbìr ì obrobka ìnformacìï. · doi:10.15407/vidbir2025.53.071
do Monte, D. M. F. M., Goldschmidt, R. R., & dos Santos, M. (2025) · Brazilian Symposium on Multimedia and the Web · doi:10.5753/webmedia.2025.16131
“…(You Only Look Once) representa uma abordagem revolucio-nária para detecção de objetos em tempo real, tratando a detecção como um problema de regressão único, onde coordenadas de bounding boxes e probabilidades de classe são preditas simultaneamente a partir de uma única avaliação da imagem [22].”
Benchmarking YOLO Variants on Aerial Object Detection with NIR-Driven Architecture
Singh, S., Pandey, D., Singh, R. P., et al. (2025) · International Conference on Reliability, Infocom Technologies and Optimization (Trends and Future Directions) (ICRITO) · doi:10.1109/icrito66076.2025.11241461
Interactive Fine-grained Few-shot Detection of Tools *
Keller, P., Strecker, L., Durchdewald, F., Graaf, F., Schnell, T., & Dillmann, R. (2025) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros60139.2025.11246465
“Popular ancestors of one-stage approaches are the Single Shot Detector (SSD) [11] and variants of the you only look once (YOLO) family [12].”
Collective Behavior Clone with Visual Attention via Neural Interaction Graph Prediction
Li, K., Ma, Z., Li, L., et al. (2025) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros60139.2025.11246098
TK-Planes: Tiered K-Planes with High Dimensional Feature Vectors for Dynamic UAV-based Scenes
Maxey, C., Choi, J., Lee, Y., Lee, H., Manocha, D., & Kwon, H. (2025) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros60139.2025.11246343 · Cited by 5
Learning to Hang Crumpled Garments with Confidence-Guided Grasping and Active Perception
Huo, S., Zhang, H., Lee, H.-Y., Zhou, P., & Navarro-Alarcón, D. (2025) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros60139.2025.11247317
“The resulting masked depth image D and the collar mask M C are then fed into the Yolo-v8 [27] model for object detection.”
Feature-aligned Fisheye Object Detection Network for Autonomous Driving
Cao, H., Sun, D., Song, R., Xia, Y., Li, X., & Knoll, A. (2025) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros60139.2025.11246605 · Cited by 1
“In contrast, one-stage methods like YOLO [12], YOLOX [13], SSD [14], and FCOS [15] offer a better balance between performance and efficiency, making them appealing for real-time applications.”
Personalized Re-identification through Unsupervised Continual Learning and Parallel Training
Rollo, F., Zunino, A., Ajoudani, A., & Kashiri, N. (2025) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros60139.2025.11246179
“It leverages the well-established YOLO framework [21] and the StrongSORT tracker [22].”
Diffusion-FS: Multimodal Free-Space Prediction via Diffusion for Autonomous Driving
Gupta, K., Stanley, T., Paul, P., Singh, A. K., & Krishna, K. M. (2025) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros60139.2025.11246466 · Cited by 2
“Unlike standard segmentation approaches [17–19], which directly classify pixels, these methods utilize learned diffusion processes to generate structured masks or extract meaningful features from the denoising steps.”
TVFET-VD:Time-Varying Formation Encircling and Tracking Control Based on Visual Detection
Yang, G., Qi, J., Wang, M., Huang, H., Peng, Y., Wu, C. H., & Ping, Y. (2025) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros60139.2025.11246132
“In terms of object detection, the proposal of object detectors based on convolutional neural networks such as YOLO [7] and SSD [8] have effectively addressed the general issues of general computer object detection, but they are not deployed in embedded robot.”
Han, L., Liu, W., Cui, J., et al. (2025) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2025.3632460
Detecting Obstacles on Railroads Using Computer Vision on UAVs
Anand, A., Krishna, N., & Vitzilaios, N. (2025) · IEEE/RJS International Conference on Intelligent RObots and Systems · doi:10.1109/iros60139.2025.11246411
“Some of the most renowned real-time detectors include You Only Look Once (YOLO) [17], known for its exceptional speed and efficiency by processing the entire image in a single pass through the network; Single Shot MultiBox Detector (SSD) [18], which balances speed and accuracy by predicting…”
Tapia, R., Ijjeh, A. A., Dios, J. R. M., et al. (2025) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros60139.2025.11247230
“To further reduce computational cost, single-stage detectors such as YOLO (You Only Look Once) [16] and SSD (Single Shot MultiBox Detector) [17] were developed.”
VisLanding: Monocular 3D Perception for UAV Safe Landing via Depth-Normal Synergy
Tan, Z., He, B., Ji, Y., & Wu, L. (2025) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros60139.2025.11247278 · Cited by 1
“Additionally, Serrano and Ban-dala (2023) [20] innovatively applied the YOLO (You Only Look Once) [21] series of object detection algorithms to safe landing zones estimation tasks.”
Graphics Detection in Document Images Using YOLO Networks
Kumar, B., Tanwar, S., Tiwari, S., et al. (2025) · International Conference on Reliability, Infocom Technologies and Optimization (Trends and Future Directions) (ICRITO) · doi:10.1109/icrito66076.2025.11241441
“You Only Look Once (YOLO) is a significant computer vision algorithm that came into existence is 2016 for object detection [1].”
Road Sight Unit: Edge Image Processing for Vehicle Trajectory and Collision Prediction
Cutileiro, J., Rosa, P. A. R., Remédios, O., et al. (2025) · doi:10.1109/nca67271.2025.00024
“We combined YOLO [11] with DeepSORT [12], which have been previously applied in road traffic environments [13].”
A steel flaw detection method: lightweight edge information extraction network
You, H., & Chen, J. (2025) · Measurement Science and Technology · doi:10.1088/1361-6501/ae2531 · Cited by 2
“Notable examples include SSD [9] and the you only look once (YOLO) series [10].”
Discriminative Image Feature Extraction for Traffic Sign Detection in Road Inspection
Wang, Y., Tang, Z., Hu, Y., Zhang, H., Low, K. H., & Zhou, X. (2025) · IEEE Transactions on Intelligent Transportation Systems · doi:10.1109/tits.2025.3632976
“In contrast, single-stage algorithms, such as the You Only Look Once (YOLO) series [8] and Single Shot MultiBox Detector (SSD) [9], omit the explicit proposal stage and perform dense prediction directly on feature maps, which typically reduces latency and makes them suited to embedded and real-time…”
YOLOv11-LWMB: A Hybrid YOLOv11 Framework for the Detection of ' Bull's-Eye ' Effect in Seismic DataPreprint
Chen, Q., Pan, L., & Jiabing, M. (2025) · Research Square · doi:10.21203/rs.3.rs-7817064/v1
Deep-Pose-Tracker: an automated behavioural analysis framework for Caenorhabditis elegansPreprint
Saha, D., Chaudhary, S., Vyas, D., Ghosh‐Roy, A., & SHARMA, R. R. (2025) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/2025.11.23.689997
Xia, Z., Zhao, Y., Xiang, H., et al. (2025) · Journal of King Saud University - Computer and Information Sciences · doi:10.1007/s44443-025-00360-4
“In contrast, single-stage detectors, such as the YOLO series (Redmon et al. 2016; Redmon and Farhadi 2018; Bochkovskiy et al. 2020; Hu et al. 2023), directly classify and regress predefined anchor boxes, offering superior computational efficiency and inference speed, making them more suitable for…”
Optimize CPU-GPU Power Allocation with Parameter and Action NetworksPreprint
Mo, Z., Xiao, Q., Shan, Z., et al. (2025) · Research Square · doi:10.21203/rs.3.rs-8015530/v1
Menghao, 杨. Y., Deqi, 蒋. J., Jiaxin, 王. W., & Xueyi, 叶. Y. (2025) · Laser & Optoelectronics Progress · doi:10.3788/lop250948
RGB-Based Staircase Detection for Quadrupedal Robots: Implementation and Analysis
Woźniak, P., Penar, P., & Bielecki, D. (2025) · Sensors · doi:10.3390/s25237247
“This was achieved using nonmaximum suppression (NMS) [42] with an intersection over union (IoU) threshold of 0.5, which means that bounding boxes with 50% or more overlap were treated as duplicates and filtered out.”
InfluentialGajoch, S., Wilk-Kołodziejczyk, D., Marcjan, Ł., Corizzo, R., Bitka, A., Małysza, M., & Skomin, G. (2025) · Materials · doi:10.3390/ma18225207
“Since its inception in [43], the architecture has undergone intensive development.”
Maia, K., Gomes, L., Vasconcelos, A., et al. (2025) · doi:10.5753/webmedia_estendido.2025.16415
“For each detected vehicle ROI (Region of Interest), a dedicated YOLOv12n [8] plate-detection model extracts the license-plate region.”
Lee, H., Si, S., Lee, K., & Youn, N. (2025) · Journal of Environmental Psychology · doi:10.1016/j.jenvp.2025.102854 · Cited by 1
Chiu, P.-W., Lin, C.-J., Chien, Y.-H., et al. (2025) · doi:10.1109/iccbe65177.2025.11256111
Adaptive Algorithms for Enhanced Night Vision in Highway Traffic Management
Tian, Z., Feng, C., Wang, Z., et al. (2025) · doi:10.1109/iccbe65177.2025.11255855
“Deep learning models, including You Only Look Once (YOLO) and faster region-based convolutional neural network (R-CNN) [5], enable real-time object detection and tracking of vehicles, pedestrians, and potential hazards.”
Lin, S.-W., Hsieh, W., Liu, Y.-S., Zhang, J., & Zhou, H.-Z. (2025) · IEEE International Conference on Computation, Big-Data and Engineering (ICCBE) · doi:10.1109/iccbe65177.2025.11255674
“The YOLO algorithm, proposed by Redmon et al. in 2016 [3], is a real-time object detection framework that outperforms two-stage detectors such as R-CNN in terms of speed by using a single-stage architecture.”
Vision-based Real-time Sign Spelling System for Cross-lingual Translation using Neural Models
Chang, Y., Lung, Y.-C., Lin, Y. C., et al. (2025) · doi:10.1109/iccbe65177.2025.11256096
“Since its introduction, it has significantly improved performance and efficiency in various applications and enabled efficient and systematic development of high-performance object detectors [2].”
Assessment of Fitness Posture Accuracy in Elderly Individuals Using Convolutional Neural Networks
Chang, C.-C., Su, K.-W., Liu, H.-P., & Hsiao, S. (2025) · 2025 IEEE International Conference on Computation, Big-Data and Engineering (ICCBE) · doi:10.1109/iccbe65177.2025.11256163
“YOLO, optimized for real-time object detection, is particularly adept at simultaneously identifying multiple body landmarks with high efficiency [10–13].”
Chen, K.-H., Chiang, P.-H., Hou, Y.-Y., Pan, Y.-J., Chen, C., & Lee, J. (2025) · IEEE International Conference on Computation, Big-Data and Engineering (ICCBE) · doi:10.1109/iccbe65177.2025.11255701
“In contrast, one-stage algorithms, including single-shot multibox detector (SSD) [6] and the YOLO series [7], eliminate the proposal generation step and directly predict object classes and bounding boxes from feature maps, significantly improving detection speed.”
Development of a Road Defect Detection and Classification System via Object Recognition Using YOLOv8
del Barrio, J. A., Yu, A., & Yumang, A. N. (2025) · doi:10.1109/icmic66299.2025.11257832
“You Only Look Once, which is an object detection model that utilizes only a single neural network to predict object classifications and bounding boxes in only one evaluation [11], has many capabilities widely explored in academic literature [12-19].”
Image Rearrangement for Fast on-Drone Tiny Object Detection
Hong, H., & Oh, C. (2025) · doi:10.1109/icmic66299.2025.11257814
“Recent advances in deep-learning–based detectors like YOLO detectors [4] have achieved remarkable accuracy and speed, and various model-compression techniques such as pruning and quantization have been proposed to reduce inference cost [5].”
Transformer-Based Object Detection in Unconstructed Environments: A ROS-Gazebo Simulation Framework
Uçar, A., Tekeş, A., & Yorulmaz, M. (2025) · International Conference on Innovations in Intelligent Systems and Applications (INISTA) · doi:10.1109/inista68122.2025.11249519
“In contrast, one-stage detectors like You Only Look Once (YOLO) [8], SSD [9], and RetinaNet [10] predict bounding boxes and class probabilities directly without a separate region proposal step.”
Video Analytics Using Deep Learning for Massive Crowd Instant Segmentation and Detection
Bhuiyan, M. R., Hassan, H., Abdullah, J., et al. (2025) · International Conference on Innovations in Intelligent Systems and Applications (INISTA) · doi:10.1109/inista68122.2025.11249555
“A number of object detectors have recently been considered accurate enough to be used in such an application, but among them, SSD [16], YOLO [15], and R-CNN [12]–[14] are the most reputable ones.”
NeuraVis: Retina Damage Detection Using Transfer Learning
Mohamed, A., Ali, S. M., Yaseen, T., et al. (2025) · International Symposium on INnovations in Intelligent SysTems and Applications · doi:10.1109/inista68122.2025.11249636
Intelligent Human Detection System in the Context of Road Safety
Scutelnicu, L.-A., & Chiriac, Ş.-M. (2025) · International Conference on Innovations in Intelligent Systems and Applications (INISTA) · doi:10.1109/inista68122.2025.11249677
“Single-stage detectors like YOLO [12] and SSD [8] are renowned for their balance of precision and processing speed, making them well-suited for real-time applications.”
Milostnaya, N. A., & Yanglyaeva, N. I. (2025) · Proceedings of Southwest State University · doi:10.21869/2223-1560-2025-29-3-193-209
Zhu, Q., Li, Y., Zhang, L., Wu, J., Shao, F., & Shen, H. (2025) · Displays · doi:10.1016/j.displa.2025.103301 · Cited by 2
An improved YOLO11 UAV toy models target detection model
Hu, Y., Pu, X., Feng, S., & Zeng, Q. (2025) · Scientific Reports · doi:10.1038/s41598-025-26601-0 · Cited by 2
Error Analysis of AI Assisted Control SystemsPreprint
Karlson, M., & Ban, H. (2025) · SSRN Electronic Journal · doi:10.2139/ssrn.5828063
Salinas, J., Diaz, M., Zavala, C., et al. (2025) · Journal of Disaster Research · doi:10.20965/jdr.2025.p0959
EXTRACTING A MUCH WIDER ABSORPTION SPECTRAL RANGE FROM A NARROWER RANGE USING TRANSFORMER MODEL
Eken, E., & Bayraklı, İ. (2025) · Konya Journal of Engineering Sciences · doi:10.36306/konjes.1662027
Qu, G., Liao, J., Liu, K., Xu, B., & Qian, Y. (2025) · Machines · doi:10.3390/machines13121107
Visual harvest: self-supervised learning for lettuce growth analysis
Simion-Constantinescu, A., & Vanschoren, J. (2025) · Acta Horticulturae · doi:10.17660/actahortic.2025.1441.29
Remote Sensing Image Retrieval Combining Global Features and Correlation-Based Local Features
Li, L., Zhou, A., & Yao, J. (2025) · IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing · doi:10.1109/jstars.2025.3639244
Dastaki, A. S., Roudposhti, K. K., Rad, M. F., et al. (2025) · doi:10.1109/ipria68579.2025.11263652
“In object tracking customization, you might adjust the size of filters W or retrain them to extract features that are most relevant to the target object (14).”
InfluentialROPH: A Robust, Optimized, and Parallelized Harris Detector with Flexible FAST-Based Pruning
Morales, A. G., Bentes, C. B., Castro, M. C., da Costa, G. A. O. P., & Tadonki, C. (2025) · International Symposium on Computer Architecture and High Performance Computing Workshops (SBAC-PADW) · doi:10.1109/sbac-padw69789.2025.00015
“A common technique in this area is the post-processing step of Non-Maximum Suppression (NMS) [13], which is employed by prominent object detection frameworks like the R-CNN family [14] and YOLO [15] to refine detections and reduce false positives.”
Gong, K., Zhang, D., Chen, M., & Wang, Y. (2025) · International Conference on Advanced Algorithms and Neural Networks (AANN) · doi:10.1109/aann66429.2025.11257605
“Frameworks for object detection, such as YOLO [3] offer notable advantages in terms of speed and efficiency.”
Evaluation of Training Strategies for Ambrosia Pollen Detection on Outdoor Images
Miličić, I., Pajević, N., Dimitrije, S., et al. (2025) · International symposium on image and signal processing and analysis/ISPA ... · doi:10.1109/ispa66905.2025.11259450
“On the other hand, one-stage models such as YOLO [6] provide faster inference.”
Nagasubramanian, D. (2025) · IEEE International Conference on Contemporary Computing and Communications (InC4) · doi:10.1109/inc465408.2025.11256484
Oe, Y., & Shoji, Y. (2025) · Lecture notes in computer science · doi:10.1007/978-3-032-11976-6_34
A precise deep learning approach for timber traceability along the forest value chain
Sandvik, Y. J., Hoseini, M., & Fischer, C. (2025) · International Journal of Forest Engineering · doi:10.1080/14942119.2025.2572938
“The You Only Look Once (YOLO) framework was first introduced by Redmon et al. (2016) and has since become one of the most popular frameworks for object detection.”
Yang, B., Cai, H. W., Sun, C., Qin, S., & Miao, G. (2025) · Arabian Journal for Science and Engineering · doi:10.1007/s13369-025-10878-4
“Representative algorithms such as Faster R-CNN [7] and the YOLO (You Only Look Once) series [8] achieve efficient object detection through end-to-end feature learning and joint optimization, providing new technical pathways for photovoltaic defect recognition.”
Wang, B., Luo, J., Lv, Y., et al. (2025) · IEEE Access · doi:10.1109/access.2025.3639311
Wang, T., Yu, L., & Ning, J. (2025) · doi:10.1109/iscipt67144.2025.11265165
“The You Only Look Once (YOLO) family has gained prominence in UAV object detection by offering an optimal balance of speed and accuracy through single-stage detection [12].”
Steel surface defect detection based on improved YOLOv8
Xia, T., & He, L. (2025) · doi:10.1109/iscipt67144.2025.11265290
“One-stage object detection algorithms [1] , such as YOLO [2] and the SSD algorithm [3] , and Two-stage object detection algorithms, such as R-CNN [4] and Faster R-CNN [5] .”
Deep Learning Based Face Detection Approaches: A Comprehensive Review
Pai, G., & Sharmila Kumari, M. (2025) · IEEE International Conference on Distributed Computing, VLSI, Electrical Circuits and Robotics (DISCOVER) · doi:10.1109/discover66922.2025.11259000
“The original YOLO [60], built on the Darknet architecture with 24 convolutional and 2 fully connected layers, was pretrained on the ImageNet dataset.”
Detection and Recognition of Infrared Small Targets Based on YOLOv5
He, Z. T., Sun, H., Zhicheng, H., et al. (2025) · IEEE Optoelectronics Global Conference (OGC) · doi:10.1109/ogc66041.2025.11262161
“Currently, YOLO (You Only Look Once) [1][2] , R-CNN (Regions with Convolutional Neural Networks) [3] , and SSD [4] (Single Shot MultiBox Detector) are three widely used object-detection frameworks, each with distinct advantages and application domains.”
HRCyolo: High-Resolution Cell Detection via Super-Resolution and Attention Mechanisms
Zhang, D., Liu, C., Liu, X., et al. (2025) · IEEE International Conference on Manipulation, Manufacturing and Measurement on the Nanoscale (3M-NANO) · doi:10.1109/3m-nano65639.2025.11261054
Qualification Considerations of Machine Learning Based Tools for Avionics System Development
Liu, C., Herencia‐Zapana, H., Nagel, S., et al. (2025) · doi:10.1109/dasc66011.2025.11257209
“• The first category contains the tools used in the ML application development lifecycle (MLDL) [16].”
The performance and improvement of one-stage networks in intelligent driving environment
Yu, J. (2025) · IEEE International Conference on Sensors, Electronics and Computer Engineering (ICSECE) · doi:10.1109/icsece65727.2025.11256979
“In 2015, Joseph et al. [9] revolutionized this domain with the You Only Look Once (YOLO) series, effectively addressing temporal constraints.”
Sarpal, S. S., Sau, S., Grover, M., et al. (2025) · doi:10.1109/mpcon66082.2025.11256724
“Numerous encodes approaches have been designed to decrease transitioning and energy usage in computer networks, with a heavy emphasis on reducing dynamic energy expenditure in bus encodes [14-15].”
Pathak, A., Agrawal, S., & Tripathi, R. (2025) · doi:10.1109/mpcon66082.2025.11256627
“Object detection models such as YOLO and Faster R-CNN have been successfully applied in warehouse environments for real-time fleet monitoring [9], [16].”
Akshaya, U., Krishnasree, A. N., Sibisiddharth, R., et al. (2025) · IEEE India Council International Subsections Conference (INDISCON) · doi:10.1109/indiscon66021.2025.11251756
“We did a benchmark of several models to decide the right fit model for the labeled SLDs. CNN [25], RCNN [26], Faster RCNN [27], and YOLO [28] models were the considered ones.”
Ardiny, H., & Beigzadeh, A. (2025) · Environmental Science and Pollution Research · doi:10.1007/s11356-025-37249-8
“However, recent advancements in artificial intelligence have led to the development of more advanced techniques, such as YOLO (Redmon et al. 2016), which offer improved accuracy and efficiency.”
Efficient SAR Vessel Detection for FPGA-Based On-Satellite Sensing
Laganier, C., Fletcher, L., Kwan, E., Walters, R., & Nockles, V. (2025) · ACM/IEEE Symposium on Edge Computing (SEC) · doi:10.1145/3769102.3772713 · Cited by 4
“YOLOv8 is a recent iteration of the YOLO architecture family [43], introducing anchor-free detection alongside improvements in both accuracy and processing speed relative to previous versions.”
Ship Target Detection in SAR Imagery Based on Band Recombination and Multi-Scale Feature Enhancement
Zhou, Y. X., Zhu, K., Guo, H., Lu, J., Gong, Z., & Liu, X. (2025) · Electronics · doi:10.3390/electronics14234728 · Cited by 1
Enhancing Automated Patient Detection and Tracking in Epilepsy Monitoring Unit Videos
Lazo, D. A. G., Wabi, I. S., Jahani, A., Gharbi, O., Nguyen, T. T., Nguyen, N., Robert, M., Sandoval, J. P. M., D’Onofrio, G., Peng, K., et al. (2025) · Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC) · doi:10.1109/embc58623.2025.11254891
“The YOLO model is a deep learning algorithm based on CNNs that performs real-time object detection and classification [10].”
Trivizakis, E., Ioannidis, G., Nikiforaki, K., Zaridis, D. I., Mylona, E., Tachos, N. S., Tsiknakis, M., Fotiadis, D. I., Regge, D., Papanikolaou, N. K., et al. (2025) · Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC) · doi:10.1109/embc58623.2025.11253440
“The You Only Look Once (YOLO) deep learning model was proposed by Redmon et al. [5] for object detection and segmentation tasks.”
Context-Aware Location De-Identification Using Denoising Diffusion
Shopon, M., & Gavrilova, M. L. (2025) · International Conference on Privacy, Security and Trust (PST) · doi:10.1109/pst65910.2025.11268887
“Mask R-CNN was chosen for this task because, unlike YOLO [21], which focuses primarily on bounding box detection, Mask R-CNN provides both object localization (bounding boxes) and segmentation (masks), making it more suitable for de-identification tasks that require precise region delineation.”
Li, A., Qu, C., & Pan, J. (2025) · Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC) · doi:10.1109/embc58623.2025.11253186
“There are some detection methods that convert EEG signals into a series of images and introduce object detection methods such as YOLO[18].”
Hara, R., Aihara, S., & Ito, T. (2025) · IEEE International Workshop on Sport, Technology and Research (STAR) · doi:10.1109/star66750.2025.11264753
“In sports-related motion analysis that uses monocular video footage, object detection methods based on the You Only Look Once (YOLO) framework [5] have been widely adopted to identify players and equipment with bounding boxes.”
YOLO-GSA: Lightweight Bearing Defect Detection Algorithm Based on Improved YOLO v5
Yang, J. M., Wu, Y., Zhan, X., et al. (2025) · doi:10.1109/icmmic66805.2025.11265233
“For example, two-stage target detection algorithms (e.g., the R-CNN [1] family) and single-stage target detection algorithms (e.g., the YOLO [2] family) have been extensively applied for defect detection.”
Reconstructing velocity profiles using scene geometry in panned and zoomed canoe sprint videos
Matthes, D., Frenzel, P., Ziegler, J., et al. (2025) · IEEE International Workshop on Sport, Technology and Research (STAR) · doi:10.1109/star66750.2025.11264798
“We employed the CNN based YOLO framework [13] for detecting buoys and athletes.”
A Federated Learning Framework for Lightweight YOLOv10 Target Detection for IoV
Du, M., Hu, X., & Bai, S. (2025) · International Congress on Image and Signal Processing, BioMedical Engineering and Informatics (CISP-BMEI) · doi:10.1109/cisp-bmei68103.2025.11259291
“YOLO (You Only Look Once) [2] is a revolutionary real-time object detection algorithm.”
YOLO-Based Whole-Slide Image Analysis System for Automated Planktonic Algae Detection
An, W. Q., Zhang, Q., Wang, J., et al. (2025) · International Congress on Image and Signal Processing, BioMedical Engineering and Informatics (CISP-BMEI) · doi:10.1109/cisp-bmei68103.2025.11259366
Image Enhancement and Recognition in Safety Supervision Scenarios Based on Multimodal Large Models
Chen, X., Wang, W., Ding, Z., & Shang, Y. (2025) · International Congress on Image and Signal Processing (CISP) · doi:10.1109/cisp-bmei68103.2025.11259279
“State-of-the-art algorithms such as Faster R-CNN [36] and YOLO [37] are trained on labeled datasets to detect PPE (e.g., protective clothing, hard hats) and unsafe behaviors Fig.”
LoRASensE: Learnable Low-Rank Acquisition in Sensors for Efficient Edge Machine Vision
Liang, Y., Yi, Z., Ma, T., et al. (2025) · IEEE/ACM International Symposium on Low Power Electronics and Design (ISLPED) · doi:10.1109/islped65674.2025.11261700
YOLOv8 with Optical Character Recognition for License Plate Recognition and Detection on a Campus
Manuel, H. R. V., Ducut, A. L. C., Macatuno, C. E. R., Serrano, K. C. S., Guintu, C. G., & Tongol, D. J. M. (2024) · IEEE International Conference on Humanoid, Nanotechnology, Information Technology, Communication and Control, Environment, and Management (HNICEM) · doi:10.1109/hnicem64917.2024.11258734
Contextual Corpus Callosum Analysis for Differentiating Early and Late Mild Cognitive Impairment
Kancharla, V. K., Bhattacharya, D., & Sinha, N. (2025) · Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC) · doi:10.1109/embc58623.2025.11254241 · Cited by 1
Transformer-Based Architectures for Real-Time Object Detection in Crowded Environments
Almalki, S. S. (2025) · International Conference on Business Analytics for Technology and Security (ICBATS) · doi:10.1109/icbats66542.2025.11258154
“YOLO introduced a real-time object detection framework that treats detection as a regression problem, directly predicting bounding boxes and class probabilities in a single forward pass [12].”
Information Compensation for Few-Shot Object Detection
Sun, Z., & Jiang, Y. (2025) · International Congress on Image and Signal Processing, BioMedical Engineering and Informatics (CISP-BMEI) · doi:10.1109/cisp-bmei68103.2025.11259202
“In recent years, with the rapid development of deep learning [19, 20], the performance of object detection models has seen significant improvement.”
FRM-YOLO: An Optimized Model for Small Object Detection in UAV Aerial Imagery
Dong, L., Liu, M., & Hu, C. (2025) · International Congress on Image and Signal Processing (CISP) · doi:10.1109/cisp-bmei68103.2025.11259409
“In deep learning, both two-stage (R-CNN [3]) and single-stage (YOLO [4]) methods perform powerfully in many applications.”
DOPDETR: An Object Detection Model with Enhanced Supervision and Parameter Sharing
Liu, M., Liu, T., Chen, L., & Zhang, J. (2025) · International Congress on Image and Signal Processing (CISP) · doi:10.1109/cisp-bmei68103.2025.11259388
“In the field of object detection, the YOLO series has continuously evolved since YOLOv1 [1], followed by YOLOv9 [2] and YOLOv10 [3].”
Poomrittigul, S., Mittong, S., Thanathornwong, B., & Suebnukarn, S. (2025) · Scientific Reports · doi:10.1038/s41598-025-30774-z
“One of the most widely used frameworks for real-time object detection is YOLO (You Only Look Once), which performs detection by dividing an image into grids and simultaneously predicting bounding boxes and class probabilities through a single neural network 9–11 .”
A Simple yet Effective Attention Acquisition Method on Marine Target Detection
Guo, L., Lyu, X., Gao, R. C., et al. (2025) · IFAC-PapersOnLine · doi:10.1016/j.ifacol.2025.11.702
LIME: High-Performance Private Inference with L ightweight M odel and Batch E ncryption
Wang, H.-C., & Wu, J. (2025) · ACM Transactions on Multimedia Computing Communications and Applications · doi:10.1145/3779221
“…to the student model; (2) the distillation loss, defined as the difference between the outputs (or logits) of the teacher and student models; and (3) the balancing parameter, which regulates the relative importance of the distillation loss and the standard cross-entropy loss during training.”
Metadata
- Authors
- Redmon, J., Divvala, S., Girshick, R., & Farhadi, A.
- Affiliations
- University of Washington (US)
- Meta (United States) (US)
- Venue
- 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)
- Publisher
- IEEE
- Published
- 2016-06
- Pages
- 779-788
- ISSN
- 1063-6919 (ISSN-L)
- OpenAlex
- W1483870316
- Cited by
- 53,803
- References
- 42
- Citations by year
- 2026: 5347 · 2025: 8912 · 2024: 8953 · 2023: 8264 · 2022: 6785 · 2021: 5741 · 2020: 4223 · 2019: 3286 · 2018: 1560 · 2017: 635 · 2016: 90
- FWCI
- 644.23
- Type
- conference paper
- Language
- en
- Open access
- green
- arxiv.org
- Locations
- IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings (publishedVersion)
- arXiv (Cornell University) (submittedVersion)
- arXiv (Cornell University)
- Zenodo (CERN European Organization for Nuclear Research) (cc-by)
- Funders
- National Science Foundation (1338054)
- Office of Naval Research (N00014-13-1-0720, N00014)
- Topics
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Advanced Neural Network Applications
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Advanced Image and Video Retrieval Techniques
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Visual Attention and Saliency Detection
- Keywords
- object detection, YOLO, real-time detection, bounding box regression, end-to-end learning, false positive reduction, domain generalization
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
The claim verdicts this report counts are experimental: their accuracy is not yet measured.
A language model drew the claims from the paper on 2026-10-10, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).
A second model labelled each claim on 2026-10-10 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).
Cited sources are checked with a language model’s help (gpt-6-sol and qwen3.8-omni-flash).
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.