Bayes, Bits and Brains
bayesbitsbrains.github.io·1d·
Discuss: Hacker News
🧠Machine Learning
Measuring Reasoning Utility in LLMs via Conditional Entropy Reduction
arxiv.org·1d
🧠Intelligence Compression
Whirlaway: Multilinear STARKs using WHIR as polynomial commitment scheme
blog.lambdaclass.com·18h
🎯Performance Proofs
Automated Species Identification via Hyper-Spectral DNA Barcoding and Machine Learning
dev.to·1h·
Discuss: DEV
🦴Database Paleontology
From Multi-Head to Latent Attention: The Evolution of Attention Mechanisms
vinithavn.medium.com·3h·
Discuss: Hacker News
⧗Information Bottleneck
What is a large language model?
proton.me·17h
💻Local LLMs
A 20-Year-Old Algorithm Can Help Us Understand Transformer Embeddings
ai.stanford.edu·2d·
Discuss: Hacker News
🧮Vector Embeddings
Unlocking Multimodal Video Transcription with Gemini
towardsdatascience.com·19h
✅Verification Codecs
The use of Claude Code in SciML repos
discourse.julialang.org·10h·
Discuss: Lobsters
📏Code Metrics
Making Operations Research More Accessible: Insights from the Rise of ML
pubsonline.informs.org·2d·
Discuss: Hacker News
🧠Machine Learning
Disentangling Latent Embeddings with Sparse Linear Concept Subspaces (SLiCS)
arxiv.org·1d
🧮Vector Embeddings
What AI chatbots are doing under the hood
gilesthomas.com·14h·
Discuss: Hacker News
🎙️Whisper
Hyper-Efficient Quantized Neural Network Pruning via Adaptive Bit-Width Allocation
dev.to·14h·
Discuss: DEV
📊Quantization
A Unified Theory of Language
arxiv.org·1d
💻Programming languages
Why One Small Parameter Can Make or Break Your Medical Image Retrieval
hackernoon.com·1d
🔍Information Retrieval
Refined bit complexity for the computation of at least onepoint per connected component of a smooth completeintersection real algebraic set
arxiv.org·1d
💎Information Crystallography
Core Concepts: Decoding Human Language
dev.to·1d·
Discuss: DEV
🤖Grammar Induction
The limits of ChatGPT: a week of interaction and AI research
marktarver.com·1d·
Discuss: Hacker News
🔲Cellular Automata
Compilation vs. vectorization, search engine edition
jpountz.github.io·1d·
Discuss: Hacker News
🚀Query Optimization
Quantization Robustness to Input Degradations for Object Detection
arxiv.org·2d
📊Quantization