Machine Learns erogol.com

Model check - DeepSeek-V3.2-Exp - Fine-Grained Sparse Attention for Efficient Long-Context LLMs

erogol.com·31w

MiMo-Audio - Scaling Speech Pre-Training to 100 Million Hours Unlocks Few-Shot Learning

erogol.com·32w

VibeVoice - Next-Token Diffusion Meets Long-Form Speech Generation

erogol.com·36w

Model check - NVIDIA Nemotron Nano 2 - An Efficient Hybrid LLM that Beats Reasoning Benchmarks

erogol.com·37w

Model check - KyutaiTTS - Streaming Text-to-Speech with Delayed Streams Modeling

erogol.com·39w

Higgs Audio V2 - Unified Audio Language Modeling at Scale

erogol.com·40w

Paper notes - Merging LLMs at Pre-training, Considering Token Probabilities at RL

erogol.com·49w

BlaGPT notes on Softpick, Canon Layers, Parallel Transformer

erogol.com·52w

A Simple Way to Switch Audio Codecs in Audio Language Models

erogol.com·75w

Solving an ML Research Problem

erogol.com·80w

Goods and Bads of Being Open Source AI Company

erogol.com·121w

Training XTTS for Voice Conversion

erogol.com·124w

XTTS v2 Notes

erogol.com·129w

XTTS v1.1 and 🐸TTS v0.19 updates

erogol.com·131w

A Review for Transformer Variants

erogol.com·135w

XTTS-v1 technical notes

erogol.com·135w

the best open-source TTS

erogol.com·137w

Notes on Fine-tuning LLMs

erogol.com·139w

Potential risks and misuses of ChatGPT, today and tomorrow.

erogol.com·140w

TODOs for Effective ML teamwork at an early-stage startup

erogol.com·187w

Log in to enable infinite scrolling