A very short, very blunt, book note
logicmatters.net·20h
⚔️Lean Theorem
Study sheds light on musicians' enhanced attention
news.mit.edu·14h·
Discuss: Hacker News
🌈Spectral Audio
Property-based testing of batch-invariant operations
mmaaz.ca·1d·
Discuss: Hacker News
🧪Property-Based Testing
PLSEMANTICSBENCH: Large Language Models As Programming Language Interpreters
arxiv.org·10h
💻Programming languages
Atomic and Saturated Models
functor.network·3d·
Discuss: Hacker News
🔢Denotational Semantics
Environment-Aware Indoor LoRaWAN Path Loss: Parametric Regression Comparisons, Shadow Fading, and Calibrated Fade Margins
arxiv.org·10h
📊Rate-Distortion Theory
From Noisy Traces to Stable Gradients: Bias-Variance Optimized Preference Optimization for Aligning Large Reasoning Models
arxiv.org·10h
🧮Constraint SMT
LLM-Guided Evolutionary Program Synthesis for Quasi-Monte Carlo Design
arxiv.org·10h
📼Tape Combinators
A Solution to the Paperclip Problem
link.springer.com·11h·
Discuss: Hacker News
🔲Cellular Automata
Speak, Edit, Repeat: High-Fidelity Voice Editing and Zero-Shot TTS with Cross-Attentive Mamba
arxiv.org·10h
🗣️CMU Pronouncing
Simulating and Understanding Deceptive Behaviors in Long-Horizon Interactions
arxiv.org·10h
✨Algebraic Effects
DiffuSpec: Unlocking Diffusion Language Models for Speculative Decoding
arxiv.org·1d
🎙️Whisper
Contrastive-SDE: Guiding Stochastic Differential Equations with Contrastive Learning for Unpaired Image-to-Image Translation
arxiv.org·10h
⧗Information Bottleneck
Probing Whisper for Dysarthric Speech in Detection and Assessment
arxiv.org·10h
🎙️Whisper
Compressed Convolutional Attention: Efficient Attention in a Compressed Latent Space
arxiv.org·10h
⧗Information Bottleneck
From Moments to Models: Graphon Mixture-Aware Mixup and Contrastive Learning
arxiv.org·10h
🧠Machine Learning
Small Language Models for Agentic Systems: A Survey of Architectures, Capabilities, and Deployment Trade offs
arxiv.org·10h
⚙️TLA+
LaDiR: Latent Diffusion Enhances LLMs for Text Reasoning
arxiv.org·10h
💻Local LLMs
Causal Abstractions, Categorically Unified
arxiv.org·10h
✨Effect Handlers
A Noise Resilient Approach for Robust Hurst Exponent Estimation
arxiv.org·10h
🧠Machine Learning