👁️ Document OCR

Text Recognition, Historical Manuscripts, Character Segmentation, Paleographic Processing

How often do movie posters put the names and faces in the wrong order?
stephenfollows.com·1d
🧠Learned Codecs
Continuous-Path 3D Printed Case is Clearly Superior
hackaday.com·2d
🌈Holographic Archives
Reflections on Matrix criticism over the last week | Lobsters
lobste.rs·3h·
Discuss: Lobsters
🚀Indie Hacking
Need Feedback: Chat with Google Search Console
asktheconsole.com·19h·
Discuss: Hacker News
📊Feed Optimization
Oracle chops well over 100 Bay Area jobs
mercurynews.com·13h
🇳🇴Norwegian Computing
You Can Now Control How Long GPT-5 Will ‘Think’
lifehacker.com·1d
🖥️Terminal Renaissance
Microsoft is getting ready to return to the office
theverge.com·17h·
Discuss: Hacker News
🔌Operating system internals
Wednesday Bytes: Blueprint, Brightspeed, 365 Data Centers, GNM
telecomramblings.com·1d
🕸️WASM Archives
AI Is Forgetting Everything – and It's Killing You
amirkabbara.substack.com·11h·
Discuss: Substack
🌍Cultural Algorithms
Numerical simulations characterize 'matrix tides' and other complex wave patterns seen in Qiantang River
phys.org·1d
🔗Tailscale
Two From Ridenour
irreal.org·2d
📟Terminal Typography
Good news or bad news?
rhollick.wordpress.com·19h
📡RSS Readers
AI model 'personalities' shape the quality of generated code
theregister.com·1d
📏Code Metrics
Sunlight-powered floating structures offer a new window into Earth's upper atmosphere
phys.org·1d
📟Terminal Physics
The GPT-5 Paradox: Genius in Thought, Gaps in Safety
dev.to·21h·
Discuss: DEV
⚡Proof Automation
Exploiting Krylov Subspace Methods for Real-Time Sparse Linear Systems in High-Dimensional Signal Processing
dev.to·3d·
Discuss: DEV
⚡SIMD Vectorization
Samsara: Valuation Compression Helps, But Execution Still Holds The Key
seekingalpha.com·2d
🔍BitFunnel
MMFformer: Multimodal Fusion Transformer Network for Depression Detection
arxiv.org·3d
🧠Machine Learning
How Effectively Can Large Language Models Connect SNP Variants and ECG Phenotypes for Cardiovascular Risk Prediction?
arxiv.org·3d
💻Local LLMs
Efficient Edge LLMs Deployment via HessianAware Quantization and CPU GPU Collaborative
arxiv.org·3d
💻Local LLMs