openreview/latex-validation

cadrille: Multi-modal CAD Reconstruction with Reinforcement Learning, ICLR desk reject, code

polyquantique/haarpy — Haarpy, a Python library for Weingarten calculus and integration of classical compact groups and ensembles

SymPy

patrick-kidger/sympy2jax — Turn SymPy expressions into trainable JAX expressions.

Attention to Mamba: A Recipe for Cross-Architecture Distillation

patrick-kidger/equinox — Elegant easy-to-use neural networks + scientific computing in JAX

klattSyn

A Speech Recognition Extension to Snack

DT2119 Calendary

olgasafonova/mediawiki-mcp-server — MCP server for MediaWiki wikis - search, read, and edit wiki content from AI assistants

Should We Still Pretrain Encoders with Masked Language Modeling?, Huggingface, training code

TMH

Strange paper titles:

Blackened Thrash Metal VS Blackened Death Metal

olvwm4 source tar

rsprouse/espsfree — BSD-licensed Entropic Signal Processing System (ESPS) released by Microsoft, repackaged and slightly updated.

Deep Learning Math Workbook

Extending papermill through entry points

huggingface/hf-mount

torchdsp/torchsig

AcousTools: A ‘Full-Stack’, Python-Based, Acoustic Holography Library, code

Introduction to Manifold Learning

54yyyu/kaggle-mcp — Kaggle-MCP: Connect Claude AI to the Kaggle API through the Model Context Protocol (MCP), enabling competition, dataset, and kernel operations through the AI interface.

googlecolab/colab-mcp — An MCP server for interacting with Google Colab

Red Hat Linux 7.0 - Appendix F. Kickstart Installations

libAudioFlux/audioFlux — A library for audio and music analysis, feature extraction.

Optimizing ASR Models with Semantic Information

VICReg: Variance-Invariance-Covariance Regularization for Self-Supervised Learning

Bridging the linguistic divide: a survey on leveraging large language models for machine translation

Aratox: a multi-dialect, multi-label arabic dataset and model benchmark for toxicity detection

Top 12 90s DEATH METAL Riffs

NVlabs/Jet-Nemotron

Toldi, audio

Whisper based Cross-Lingual Phoneme Recognition between Vietnamese and English, arXiv

VICReg Pytorch (official), Jax

facebookresearch/sam2

IDEA-Research/DWPose

lllyasviel/ControlNet

How to Generate Text in One Step

antirez/qwen-asr — C inference for Qwen3-ASR 0.6b and 1.7b transcriptions models

antirez/sds — Simple Dynamic Strings library for C

IEMOCAP, Kaggle, HF

ljuvela/GlotNet

ljuvela/SourceFilterNeuralFormants — Neural formant synthesis using differtiable resonant filters and source-filter model structure.

HiFi-Glot: High-Fidelity Neural Formant Synthesis with Differentiable Resonant Filters, code, model

deepinsight/insightface

LRE Journal - template, overleaf (broken link); Wayback

Submission guidelines

The NeWMe corpus: a gold standard corpus for the study of word meaning negotiation

The Kestrel TTS text normalization system

Multilingual text analysis for text-to-speech synthesis

How To Become a Mathematical Genius

Personal Exceptionalism: why knowing you’re the exception to the rule matters

  1. Take on insane goals (that stretch beyond skill/experience level)
  2. Break the rules
  3. Be resilient
  4. Push themselves
  5. Upside thinking

Laminate, Immodium

Spectral Embedding and Laplacian Eigenmaps

Blending Supervised and Reinforcement Fine-Tuning with Prefix Sampling, code

Streaming Speech-to-Text Translation with a SpeechLLM

NVlabs/Sana — Efficient High-Resolution Image Synthesis with Linear Diffusion Transformer. — Code is open, models are not.

Wikipedia volunteers spent years cataloging AI tells. Now there’s a plugin to avoid them., code

ericcurtin/wayoa

poloclub/transformer-explainer

jamiepine/voicebox

ysharma3501/LuxTTS — LuxTTS is an lightweight zipvoice based text-to-speech model designed for high quality voice cloning and realistic generation at speeds exceeding 150x realtime.

k2-fsa/ZipVoice — Fast and High-Quality Zero-Shot Text-to-Speech with Flow Matching

the largest cocaine-mixing factory ever found in the history of the State