Learning languages with the help of algorithms
johndcook.com·1d·
Discuss: Hacker News
🧮Kolmogorov Bounds
Parallelizable Feynman-Kac Models for Universal Probabilistic Programming
arxiv.org·1d
🧮Kolmogorov Bounds
Latent trees
languagelog.ldc.upenn.edu·7h
🤖Grammar Induction
Thermodynamically Optimized Data Batching via Information-Theoretic Reservoir Sampling in HBM
dev.to·13h·
Discuss: DEV
🌊Streaming Compression
Worth Reading 091925
rule11.tech·4h
🕸️Mesh Networking
An Interactive Guide to 4 Fundamental Computer Vision Tasks Using Transformers
towardsdatascience.com·5h
🤖Advanced OCR
Adding LLMs to the psycholinguistic norming toolbox: A practical guide to getting the most out of human ratings
arxiv.org·15h
🧠Intelligence Compression
Language Models Pack Billions of Concepts into 12,000 Dimensions
nickyoder.com·4d·
Discuss: Lobsters, Hacker News
🧠Machine Learning
Disproving the Feasibility of Learned Confidence Calibration Under Binary Supervision: An Information-Theoretic Impossibility
arxiv.org·15h
🧠Machine Learning
Building sub-100ms autocompletion for JetBrains IDEs
blog.sweep.dev·2h·
Discuss: Hacker News
🔨Compilers
Emergent Alignment via Competition
arxiv.org·15h
🔲Cellular Automata
A Universal Banach--Bregman Framework for Stochastic Iterations: Unifying Stochastic Mirror Descent, Learning and LLM Training
arxiv.org·1d
🧠Machine Learning
Joint Scheduling and Multiflow Maximization in Wireless Networks
arxiv.org·15h
📡Network Calculus
On the complexity of unique quantum witnesses and quantum approximate counting
arxiv.org·15h
⚛️Quantum Circuits
Should GPUs Make Free Trade Agreements?
doubleword.ai·2h·
Discuss: Hacker News
🌊Streaming Algorithms
Combining Minkowski and Chebyshev: New distance proposal and survey of distance metrics using k-nearest neighbours classifier
arxiv.org·15h
🌀Differential Geometry
Smarter Pixels: Turbocharging Visual AI with Semantic Compression by Arvind Sundararajan
dev.to·1d·
Discuss: DEV
🧠Neural Compression
Large Language Model probabilities cannot distinguish between possible and impossible language
arxiv.org·15h
💻Programming languages
Graph-Augmented Hybrid Retrieval and Multi-Stage Re-ranking: A Framework for High-Fidelity Chunk Retrieval in RAG Systems
dev.to·1d·
Discuss: DEV
🎯Retrieval Systems
Such a Classic
blog.hermesloom.org·2d·
Discuss: Hacker News
🧠Intelligence Compression