When Models Manipulate Manifolds: The Geometry of a Counting Task
transformer-circuits.pub·53m·
Discuss: Hacker News
🌀Differential Geometry
Flag this post
Time-Embedded Algorithm Unrolling for Computational MRI
arxiv.org·14h
🧠Machine Learning
Flag this post
A lifting week
blog.mitrichev.ch·23h·
🎯Performance Proofs
Flag this post
Unlock AI's Speed Potential: Bitwidth-Aware Logarithmic Arithmetic by Arvind Sundararajan
dev.to·10h·
Discuss: DEV
🔢Bitwise Algorithms
Flag this post
A deep dive into BPF LPM trie performance and optimization
blog.cloudflare.com·1h·
Discuss: Hacker News
🌳Trie Structures
Flag this post
Denosing Images of Cats and Dogs with Autoencoders
mayberay.bearblog.dev·5h·
Discuss: Hacker News
🧠Neural Codecs
Flag this post
Does subgroup/wave size matter?
gfxstrand.net·9h·
Discuss: Hacker News
💎Information Crystallography
Flag this post
LLM hallucinations are compression failures, and we can detect them
leon168689.substack.com·1d·
Discuss: Substack
🧮Kolmogorov Bounds
Flag this post
Exploiting the Potential of Linearity in Automatic Differentiation and Computational Cryptography
arxiv.org·14h
👑Coq Tactics
Flag this post
Forward-Backward Binarization
arxiv.org·14h
Bidirectional Programming
Flag this post
Volumetric Non-Invasive Cardiac Mapping for Accessible Global Arrhythmia Characterization
arxiv.org·14h
💾Floppy Imaging
Flag this post
Topology-Aware Hybrid Wi-Fi/BLE Fingerprinting via Evidence-Theoretic Fusion and Persistent Homology
arxiv.org·14h
📡Bluetooth Archaeology
Flag this post
Predictability of Complex Systems
arxiv.org·14h
🔲Cellular Automata
Flag this post
I Built a SAT Solver Inspired by Quantum Field Theory and It Works
sethuiyer.github.io·1d·
Discuss: Hacker News
⚛️Information Physics
Flag this post
Nonlinear Dimensionality Reduction Techniques for Bayesian Optimization
arxiv.org·1d
🧠Machine Learning
Flag this post
HaTT: Hadamard avoiding TT recompression
arxiv.org·14h
🕸️Tensor Networks
Flag this post
Is Zadeh's Least-Entered Pivot Rule Exponential?
arxiv.org·14h
🎯Proof Tactics
Flag this post
Optimizing LLM Context for Vulnerability Scanning
blog.fraim.dev·1d·
Discuss: Hacker News
🧪Binary Fuzzing
Flag this post
FrugalPrompt: Reducing Contextual Overhead in Large Language Models via Token Attribution
arxiv.org·14h
⚙️Compression Benchmarking
Flag this post