Your Transformer is Secretly an EOT Solver
elonlit.com·1d·
Discuss: Hacker News
🪜Recursive Descent
Flag this post
GIR-Bench: Versatile Benchmark for Generating Images with Reasoning
paperium.net·14h·
Discuss: DEV
🗺️Region Polymorphism
Flag this post
From Lossy to Lossless Reasoning
manidoraisamy.com·12h·
Discuss: Hacker News
🪜Recursive Descent
Flag this post
A Short Introduction to Categorical Logic
youtube.com·2d
⚖️Inference Rules
Flag this post
AI Experiments: Fast Inference with Groq and Third-Party Tools with Kimi K2 in TypingMind
macstories.net·1d
🌱Minimal Interpreters
Flag this post
Cross-Platform Evaluation of Reasoning Capabilities in Foundation Models
arxiv.org·1d
🔶Lattice Theory
Flag this post
From Scripts to Scale: Python, Mypy, and the Rise of Static Typing
simplethread.com·1d·
Discuss: Hacker News
🌊Gradual Typing
Flag this post
GHC 9.14.1-rc1 is now available
discourse.haskell.org·21h·
📏Linear Haskell
Flag this post
Fungus: The Befunge CPU(2015)
bedroomlan.org·4h·
Discuss: Hacker News
🌳B+ Trees
Flag this post
Making Sense of Lambda Calculus 6: Recurring Problems
aartaka.me·2d·
λLambda Calculus
Flag this post
Introducing Project Telos: Modeling, Measuring, and Intervening on Goal-directed Behavior in AI Systems
lesswrong.com·21h
🎭Erlang OTP
Flag this post
Building a Visual Diff System for AI Edits (Like Git Blame for LLM Changes)
news.ycombinator.com·10h·
Discuss: Hacker News
🌊Gradual Effects
Flag this post
Magisterium
isaacbowen.com·12h·
Discuss: Hacker News
💫Effect Systems
Flag this post
Master Rust Pattern Matching: Build Safer, More Expressive Code with Advanced Techniques
dev.to·1d·
Discuss: DEV
🎯Pattern Matching
Flag this post
Language Models are Injective and Hence Invertible - GLADIA Research
arxiv.org·3d
🪜Recursive Descent
Flag this post
PORTool: Tool-Use LLM Training with Rewarded Tree
arxiv.org·1d
🪜Recursive Descent
Flag this post
TinyML is the most impressive piece of software you can run on any ESP32
xda-developers.com·20h
💬Smalltalk VMs
Flag this post
Show HN: Everything it took to run an LLM at 10k tok/s on H200s
relace.ai·2d·
Discuss: Hacker News
Gleam
Flag this post