1 53

Markus Strazds PRO

MStrazds

AI & ML interests

None yet

Recent Activity

liked a model about 1 month ago

fastino/gliner2-base-v1

liked a model about 1 month ago

onnx-community/gliner_small-v2.1

liked a model about 1 month ago

urchade/gliner_small-v2.1

View all activity

Organizations

liked 3 models about 1 month ago

liked a model about 2 months ago

jinaai/jina-embeddings-v5-text-nano

Feature Extraction • 0.2B • Updated Apr 15 • 331k • 72

liked a model 2 months ago

Qwen/Qwen3.5-9B

Image-Text-to-Text • 10B • Updated Mar 2 • 7.84M • • 1.49k

liked 3 models 3 months ago

Qwen/Qwen3-Embedding-0.6B

Feature Extraction • 0.6B • Updated Apr 20 • 7.65M • • 1.04k

nomic-ai/nomic-embed-text-v1.5

Qwen/Qwen3.5-35B-A3B

Image-Text-to-Text • 36B • Updated Apr 24 • 3.01M • • 1.44k

liked a Space 3 months ago

Invisible Watermark Against Unauthorized AI Training — Text, Image & Video Protection

⚡

One embed. Four invisible layers. 34 attacks defeated.

reacted to SeaWolf-AI's post with ❤️ 3 months ago

Post

2746

AI Is Training on Your Content Without Permission — Fight Back with Invisible Watermarks

FINAL-Bench/security-scan

Most generative AI training data is crawled without consent. Your text gets summarized, images reprocessed, videos clipped — with no way to prove you're the original creator. Existing watermarks are either visible or wiped out by a single AI preprocessing pass.

Detect Before, Track After

Pre-embed — Detect theft without any watermark. Text plagiarism check, image similarity analysis (perceptual hash, SSIM, color histogram, feature matching), and video temporal matching catch copies, edits, and excerpts.

Post-embed — Embed invisible multi-layer watermarks. If one layer is destroyed, others survive independently. Even full removal leaves forensic traces as evidence.

Text: 4 Independent Layers

Four mechanisms work simultaneously: zero-width Unicode characters at morpheme/word boundaries (Korean Kiwi + English NLP), style fingerprinting via synonym-ending-connective substitution, SHA-256 timestamped evidence packages, and punctuation-anchored micro-marks. Each layer uses a different Unicode category, so attacks on one cannot eliminate the others. Full bilingual support, zero readability impact.

34-Attack Defense

7 categories, 34 attacks simulated: Unicode normalization, invisible character removal, homoglyph substitution (9,619 confusables), and AI rewriting. Each scored on Signal (watermark survival) + Trace (forensic evidence of attack) — proving deliberate removal even when watermarks are destroyed.

Image & Video

Images: DCT frequency-domain watermarks surviving JPEG compression and resize. Videos: keyframe watermarking with temporal propagation and majority-vote extraction. Both support pre-embed similarity detection.

Who Is This For

Creators, rights holders needing legal evidence, media companies, and organizations tracking document leaks. Korean/English bilingual, open source, Gradio-based.