Hugging Face
Models
Datasets
Spaces
Buckets
new
Docs
Enterprise
Pricing
Log In
Sign Up
226.5
TFLOPS
128
12
138
Fatih C. Akyon
fcakyon
Follow
Mustafaege's profile picture
onuralpszr's profile picture
Alptekinege's profile picture
56 followers
·
26 following
https://linktr.ee/fcakyon
fcakyon
fcakyon
fcakyon
AI & ML interests
VLM, Agentic Dataset Curation @ultralytics @metu
Recent Activity
authored
a paper
about 16 hours ago
SenBen: Sensitive Scene Graphs for Explainable Content Moderation
posted
an
update
about 19 hours ago
Let me introduce you to our CVPR 2026 paper! Today's content moderation systems give you a label: safe or unsafe. They don't tell you what triggered the decision, who is involved, or where in the image it happens. That opacity hurts auditing, breaks adaptation across platforms, and frustrates the human review that responsible deployment demands. We built SenBen to fix this: the first large-scale scene graph benchmark designed specifically for sensitive content moderation: - 13,999 annotated frames from 157 movies - Visual Genome style scene graphs with bounding boxes, attributes, and predicates - Affective state attributes (pain, fear, aggression, distress) so the model captures not just what is in the frame, but what it means - 16 safety tags across 5 categories, the broadest taxonomy of any dataset of this kind A small model that beats much bigger ones: We distilled a frontier VLM into a compact 241M parameter student built on Florence-2. On grounded scene graph metrics, the 241M student beats every evaluated VLM except Gemini, and every commercial safety API. It also wins on object detection and captioning across the entire model zoo. It runs at 733 ms per frame on 1.2 GB VRAM, which is 7.6 times faster than the next-best local VLM at zero per-frame cost. The whole benchmark, from dataset creation through all baseline evaluations, is reproducible for under $350. Project: https://senben.kim/ Paper: https://huggingface.co/papers/2604.08819 Dataset: https://huggingface.co/datasets/fcakyon/senben Code (soon): https://github.com/fcakyon/senben
updated
a dataset
9 days ago
fcakyon/senben
View all activity
Organizations
fcakyon
's models
16
Sort: Recently updated
fcakyon/florence-2-base
Image-Text-to-Text
•
Updated
May 22, 2025
•
9
fcakyon/timesformer-large-finetuned-k600
Video Classification
•
0.1B
•
Updated
Mar 17, 2025
•
5
•
1
fcakyon/timesformer-hr-finetuned-k400
Video Classification
•
0.1B
•
Updated
Feb 12, 2025
•
4
fcakyon/timesformer-hr-finetuned-k600
Video Classification
•
0.1B
•
Updated
Jan 22, 2025
•
3
fcakyon/timesformer-base-finetuned-k400
Video Classification
•
0.1B
•
Updated
Jan 22, 2025
•
7
fcakyon/timesformer-large-finetuned-k400
Video Classification
•
0.1B
•
Updated
Jan 22, 2025
•
94
•
1
fcakyon/test-model
Image Segmentation
•
Updated
Jan 15, 2023
•
13
fcakyon/yolov5n-cls-v7.0
Image Classification
•
Updated
Dec 20, 2022
•
12
•
2
fcakyon/yolov5n-seg-v7.0
Updated
Dec 20, 2022
•
10
fcakyon/yolov5n-v7.0
Object Detection
•
Updated
Dec 20, 2022
•
5
•
3
fcakyon/yolov5s-v7.0
Object Detection
•
Updated
Dec 20, 2022
•
42
•
13
fcakyon/timesformer-large-finetuned-ssv2
Video Classification
•
Updated
Dec 10, 2022
•
7
fcakyon/timesformer-base-finetuned-ssv2
Video Classification
•
Updated
Dec 10, 2022
•
5
fcakyon/timesformer-base-finetuned-k600
Video Classification
•
Updated
Dec 10, 2022
•
3
fcakyon/timesformer-hr-finetuned-ssv2
Video Classification
•
Updated
Dec 10, 2022
•
5
fcakyon/mmdet-yolox-tiny
Object Detection
•
Updated
Dec 5, 2022
•
3