This model is designed to detect question regions in exam booklets, worksheets, and PDF page renders. It works best on dense educational pages with clear question boundaries.

Release notes

This release checkpoint is a cleaned fine-tune of a previously trained YOLO26M run, finalized without mosaic augmentation.

Compared with the previously published HF YOLO11M baseline, this release improves held-out mAP50-95 by +0.022 on the old test set and +0.057 on the combined test set.

Test results

Old held-out test set

Model	Size	Precision	Recall	mAP50	mAP50-95
YOLO11M Question Segmentation v1	116 MB / 20.1M params	0.962	0.980	0.979	0.898
YOLO26M Question Segmentation v2	168 MB / 21.8M params	0.990	0.982	0.988	0.920
YOLO26S Question Segmentation v2	20 MB / 9.9M params	0.988	0.983	0.990	0.923
YOLO26N Question Segmentation v2	5.3 MB / 2.5M params	0.988	0.983	0.989	0.924

Combined held-out test set

Model	Size	Precision	Recall	mAP50	mAP50-95
YOLO11M Question Segmentation v1	116 MB / 20.1M params	0.962	0.973	0.984	0.900
YOLO26M Question Segmentation v2	168 MB / 21.8M params	0.987	0.978	0.992	0.957
YOLO26S Question Segmentation v2	20 MB / 9.9M params	0.988	0.991	0.993	0.963
YOLO26N Question Segmentation v2	5.3 MB / 2.5M params	0.991	0.988	0.993	0.962

Benchmark graphs

Confidence guidance

From the confidence sweep used in this release:

Best benchmark confidence for YOLO26S and YOLO26N is conf=0.001 (highest mAP50-95 on both held-out test sets).
For practical inference, start with conf=0.25; use conf=0.20 if you want fewer missed borderline boxes.

Qualitative examples

The newer YOLO26 checkpoints improve most clearly in two cases:

Less false positives on pages with no questions.
Better recovery of harder placements such as one-row questions and questions that occupy an entire column.

Example no-question page: GT + YOLO11M Question Segmentation v1 + YOLO26M Question Segmentation v2 + YOLO26S Question Segmentation v2 + YOLO26N Question Segmentation v2. The YOLO11M panel uses the earlier low-threshold prediction pass and shows two false positives; the YOLO26 checkpoints stay empty on the same page.

Usage

python

from huggingface_hub import hf_hub_download
from ultralytics import YOLO

weights = hf_hub_download(repo_id="erayyapagci/yolo26m-question-segmentation-v2", filename="best.pt")
model = YOLO(weights)
results = model("page.png", imgsz=1280, conf=0.25)

Direct URL loading also works:

python

from ultralytics import YOLO

model = YOLO("https://huggingface.co/erayyapagci/yolo26m-question-segmentation-v2/resolve/main/best.pt")
results = model("page.png", imgsz=1280, conf=0.25)

Data note

Some of the training data comes from public Roboflow projects used in earlier versions of this question-segmentation pipeline:

PDF Soru Cikarma (tanimazsinu): Link
WholeQuestionDetection (Gazi University): Link
ExamBuddy (ExamBuddy): Link
Questions (Terry Li): Link
Question Parsing from Document (Sefa): Link
Question Dedector (Nur Etinkaya): Link
Sorukes (Sorualgilama): Link
Question Detection (Cognizen): Link
Questions2 (Fiver): Link
Question-New (Question): Link

License

This release is published under CC-BY-4.0.