Channel Simulation and Distributed Compression with Ensemble Rejection Sampling
arxiv.orgยท12h
โง—Information Bottleneck
Three Solutions to Nondeterminism in AI
blog.hellas.aiยท4hยท
Discuss: Hacker News
๐ŸŽฏPerformance Proofs
Efficient and accurate search in petabase-scale sequence repositories
nature.comยท38m
๐Ÿ”„Burrows-Wheeler
2D Phase Space Tomography: Validating MENT-Flow Against MENT Using 1D Projections
hackernoon.comยท1d
๐Ÿ”—Tailscale
Introducing OpenZL: An Open Source Format-Aware Compression Framework
engineering.fb.comยท2dยท
โšกModern Compression
SSDD: Single-Step Diffusion Decoder for Efficient Image Tokenization
github.comยท21hยท
Discuss: Hacker News
โšกModern Compression
H1B-KV: Hybrid One-Bit Caches for Memory-Efficient Large Language Model Inference
arxiv.orgยท12h
๐Ÿ’จCache Optimization
The Mondrian introduction to functional optics
marcosh.github.ioยท1dยท
โŸทBidirectional Programming
Quantum capacity amplification via privacy
arxiv.orgยท1d
โš›๏ธQuantum Compression
ECLipsE-Gen-Local: Efficient Compositional Local Lipschitz Estimates for Deep Neural Networks
arxiv.orgยท12h
๐ŸŒ€Riemannian Computing
JEPAs Unveiled: How Your AI Implicitly Knows Your Data's Secrets
dev.toยท8hยท
Discuss: DEV
๐Ÿ”Vector Forensics
Partial Information Decomposition via Normalizing Flows in Latent Gaussian Distributions
arxiv.orgยท1d
๐Ÿง Machine Learning
Fast Matrix Multiply on an Apple GPU
percisely.xyzยท19hยท
Discuss: Hacker News
โšกSIMD Vectorization
Quantifying the Accuracy-Interpretability Trade-Off in Concept-Based Sidechannel Models
arxiv.orgยท12h
๐Ÿง Machine Learning
Less Is More: Recursive Reasoning with Tiny Networks
github.comยท1hยท
Discuss: Hacker News
๐Ÿ“ŠQuantization
Experimental Proof: MENT-Flow Achieves Consistency with MENT Solutions for Beam Reconstruction
hackernoon.comยท1d
๐Ÿง Learned Codecs
Optimized Minimal 4D Gaussian Splatting
arxiv.orgยท1d
๐Ÿ“Projective Geometry
Parameterized Complexity of Temporal Connected Components: Treewidth and k-Path Graphs
arxiv.orgยท12h
๐ŸŽจGraph Coloring