Mishearings: Hacking a tiny ASR model to write Dadaist poetry
evanking.io·2d·
Discuss: Hacker News
🎙️Whisper
READER: Retrieval-Assisted Drafter for Efficient LLM Inference
arxiv.org·3h
🔄Burrows-Wheeler
Subsampling Factorization Machine Annealing
arxiv.org·3h
🌊Streaming Algorithms
Technical Report: Full-Stack Fine-Tuning for the Q Programming Language
arxiv.org·1d
💻Local LLMs
Recent cross-research on LLM and RL on ArXiv
github.com·2d·
Discuss: Hacker News
🔲Cellular Automata
BiasGym: Fantastic Biases and How to Find (and Remove) Them
arxiv.org·3h
🧠Intelligence Compression
Stationarity Exploration for Multivariate Time Series Forecasting
arxiv.org·3h
📈Time Series
Scaled-Dot-Product Attention as One-Sided Entropic Optimal Transport
arxiv.org·3h
🧠Machine Learning
Audio-Thinker: Guiding Audio Language Model When and How to Think via Reinforcement Learning
arxiv.org·1d
🎵Audio ML
KFFocus: Highlighting Keyframes for Enhanced Video Understanding
arxiv.org·3h
📊Rate-Distortion Theory
First Ask Then Answer: A Framework Design for AI Dialogue Based on Supplementary Questioning with Large Language Models
arxiv.org·3h
🎙️Whisper
Programming a Language
nrinaudo.github.io·1d·
Discuss: Hacker News
🔗Concatenative Programming
ReQuestNet: A Foundational Learning model for Channel Estimation
arxiv.org·3h
🧠Machine Learning
Fuzzy-Pattern Tsetlin Machine
arxiv.org·3h
🔤Morris-Pratt
Rethinking Tokenization for Rich Morphology: The Dominance of Unigram over BPE and Morphological Alignment
arxiv.org·3h
📋Document Grammar
From Product Hilbert Spaces to the Generalized Koopman Operator and the Nonlinear Fundamental Lemma
arxiv.org·1d
🕸️Tensor Networks
A Risk Taxonomy and Reflection Tool for Large Language Model Adoption in Public Health
arxiv.org·3h
✨Effect Handlers
Quantifying Marine Ecosystem Resilience via Hyperdimensional Spectral Analysis of Bioacoustic Data
dev.to·1d·
Discuss: DEV
📡Frequency Archaeology
When and how can inexact generative models still sample from the data manifold?
arxiv.org·1d
🌀Differential Geometry
Exploring Causal Effect of Social Bias on Faithfulness Hallucinations in Large Language Models
arxiv.org·1d
💻Programming languages