🔄 In a Training Loop

Manuel Romero PRO

mrm8488

·

https://mrm8488.github.io

AI & ML interests

#AI Research and Democratization. NLP/NLG 🤗

Recent Activity

liked a dataset about 7 hours ago

rogue-security/prompt-injections-benchmark

upvoted a collection 11 days ago

upvoted a collection 11 days ago

View all activity

Organizations

upvoted 2 collections 11 days ago

MobileMoE

a family of on-device MoE language models with sub-billion active parameters (0.3-0.9B active and 1.3-5.3B total) that establish a new Pareto frontier • 11 items • Updated 17 days ago • 28

NeoMME

Meet NeoMME: a family of 260M and 800M Multimodal-Native Multilingual Encoders • 12 items • Updated 14 days ago • 33

upvoted 3 articles about 1 month ago

Article

Multi-Vector (Late Interaction) Embedding Models with Sentence Transformers

+1

tomaarsen, NohTow, raphaelsty

•

about 1 month ago

• 114

Article

I fine-tuned a model for free from one prompt, with TRL and the Google Colab CLI

sergiopaniego

•

Jun 15

• 8

Article

mDenseOn with the mLateOn: Open Multilingual, Long-Context, and Code Retrieval Models

lightonai

•

Jul 30

• 38

upvoted a collection 2 months ago

Bonsai 27B

10 items • Updated Jul 15 • 234

upvoted an article 2 months ago

Article

Distillation in 2026 (so far): which frontier models use it and how

sergiopaniego

•

Jul 8

• 21

upvoted a collection 5 months ago

DeepSeek-V4

10 items • Updated 8 days ago • 910

upvoted a changelog 5 months ago

Hugging Face Changelog

Spaces agents.md for your coding agents

Apr 17

• 348

upvoted 2 articles 5 months ago

Article

The PR you would have opened yourself

pcuenq, awni

•

Apr 16

• 73

Article

BidirLM: Turning Generative LLMs into the Best Open-Source Omnimodal Encoders

Nicolas-BZRD

•

Apr 7

• 28

upvoted a paper 5 months ago

BidirLM: From Text to Omnimodal Bidirectional Encoders by Adapting and Composing Causal LLMs

Paper • 2604.02045 • Published Apr 2 • 40

upvoted a collection 6 months ago

Gemma 4

Our most intelligent open models to date • 16 items • Updated 28 days ago • 1.13k

upvoted 6 papers 6 months ago

Trace2Skill: Distill Trajectory-Local Lessons into Transferable Agent Skills

Paper • 2603.25158 • Published Mar 26 • 56

REAP the Experts: Why Pruning Prevails for One-Shot MoE compression

Paper • 2510.13999 • Published Oct 15, 2025 • 23

OpenResearcher: A Fully Open Pipeline for Long-Horizon Deep Research Trajectory Synthesis

Paper • 2603.20278 • Published Mar 17 • 103

MinerU-Diffusion: Rethinking Document OCR as Inverse Rendering via Diffusion Decoding

Paper • 2603.22458 • Published Mar 23 • 139

Recursive Language Models Meet Uncertainty: The Surprising Effectiveness of Self-Reflective Program Search for Long Context

Paper • 2603.15653 • Published Mar 7 • 13

Nemotron-Cascade 2: Post-Training LLMs with Cascade RL and Multi-Domain On-Policy Distillation

Paper • 2603.19220 • Published Mar 19 • 70

upvoted an article 6 months ago

Article

LoRA Fine-Tuning BitNet b1.58 LLMs on Heterogeneous Edge GPUs via QVAC Fabric

qvac

•

Mar 17

• 19