FineBooks: are open OCR models good enough to unlock historical knowledge?
• 27
AI & ML interests
None defined yet.
Recent Activity
View all activity
Articles
Board: add Surya OCR 2.5 (bhl-ocr-eval#14, 79f1f2b)
#7 opened 1 day ago
by 
davanstrien
davanstrien
updated a
bucket 1 day ago
davanstrien
published a
bucket 1 day ago
Add MiniCPM-OCR-1.0 (rank 8, 0.0368), from bhl-ocr-eval db3c0ee
#6 opened 3 days ago
by davanstrien
Add Unlimited-OCR-2 (rank 6, 0.0329), from bhl-ocr-eval 0bda3b9
#5 opened 4 days ago
by davanstrien
Apples to apples
2
#3 opened 6 days ago
by 
johnlockejrr
Add OXR-1.0 and OvisOCR3 (bhl-ocr-eval #10)
#4 opened 5 days ago
by davanstrien
Add nvidia/Nemotron-3.5-Parse (bhl-ocr-eval #9, b2cf5ff)
#2 opened 6 days ago
by davanstrien
Bake: add Chandra-OCR-2, LightOnOCR-3 (4B, 0.8B), Indra OCR (bhl-ocr-eval c3fb140)
#1 opened 8 days ago
by davanstrien
davanstrien
updated a
Space about 2 months ago
storytracer
updated a
dataset 3 months ago
davanstrien
published an article 3 months ago
Article
FineBooks: are open OCR models good enough to unlock historical knowledge?
finebooks
• • 27 davanstrien
published a
Space 3 months ago
storytracer
updated a
Space 3 months ago
storytracer
published a
Space 3 months ago
storytracer
published a
dataset 4 months ago
storytracer
updated a
dataset 4 months ago
storytracer
published a
dataset 4 months ago
davanstrien
posted an update about 1 year ago
Post
3,711
I fine-tuned a smol VLM to generate specialized art history metadata!
https://huggingface.co/davanstrien/iconclass-vlm: Qwen2.5-VL-3B trained using SFT to generate ICONCLASS codes (think Dewey Decimal for art!)
Trained with TRL + HF Jobs - single UV script, no GPU needed!
Space to explore predictions on a test set: davanstrien/iconclass-predictions
Blog soon!
https://huggingface.co/davanstrien/iconclass-vlm: Qwen2.5-VL-3B trained using SFT to generate ICONCLASS codes (think Dewey Decimal for art!)
Trained with TRL + HF Jobs - single UV script, no GPU needed!
Space to explore predictions on a test set: davanstrien/iconclass-predictions
Blog soon!
- 1 reply
davanstrien
posted an update over 1 year ago
Post
3,786
Inspired by Hugging Face's official MCP server, I've developed a complementary tool that exposes my semantic search API to enhance discovery across the HF platform.
Key capabilities:
- AI-powered semantic search for models and datasets
- Parameter count analysis via safetensors metadata
- Trending content discovery
- Find similar models/datasets functionality
- 11 tools total for enhanced ecosystem navigation
The semantic search goes beyond simple keyword matching, understanding context and relationships between different models and datasets.
Example query: "Find around 10 reasoning Hugging Face datasets published in 2025 focusing on topics other than maths and science. Show a link and a short summary for each dataset." (results in video!)
https://github.com/davanstrien/hub-semantic-search-mcp
Key capabilities:
- AI-powered semantic search for models and datasets
- Parameter count analysis via safetensors metadata
- Trending content discovery
- Find similar models/datasets functionality
- 11 tools total for enhanced ecosystem navigation
The semantic search goes beyond simple keyword matching, understanding context and relationships between different models and datasets.
Example query: "Find around 10 reasoning Hugging Face datasets published in 2025 focusing on topics other than maths and science. Show a link and a short summary for each dataset." (results in video!)
https://github.com/davanstrien/hub-semantic-search-mcp
- 1 reply