Manuel Romero's picture

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

BidirLM/BidirLM-Omni-Contrastive

upvoted an article about 7 hours ago

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

upvoted a paper about 7 hours ago

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

View all activity

Organizations

upvoted an article about 7 hours ago

Article

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

1 day ago

•

18

upvoted a paper about 7 hours ago

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

Paper • 2604.02045 • Published 7 days ago • 19

upvoted a collection 6 days ago

Gemma 4

8 items • Updated 7 days ago • 500

upvoted a paper 10 days ago

Trace2Skill: Distill Trajectory-Local Lessons into Transferable Agent Skills

Paper • 2603.25158 • Published 14 days ago • 49

upvoted a paper 12 days ago

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

Paper • 2510.13999 • Published Oct 15, 2025 • 19

upvoted 2 papers 15 days ago

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

Paper • 2603.20278 • Published 22 days ago • 93

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

Paper • 2603.22458 • Published 16 days ago • 134

upvoted 2 papers 19 days ago

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

Paper • 2603.15653 • Published Mar 7 • 12

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

Paper • 2603.19220 • Published 20 days ago • 66

upvoted an article 22 days ago

Article

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

23 days ago

•

15

upvoted a paper 27 days ago

Decoupling Reasoning and Confidence: Resurrecting Calibration in Reinforcement Learning from Verifiable Rewards

Paper • 2603.09117 • Published about 1 month ago • 9

upvoted a collection 27 days ago

Qwen3.5-text-only

Text-only versions of Qwen-3.5 without the vision encoders for a smaller memory and storage footprint. • 4 items • Updated about 16 hours ago • 14

upvoted an article 29 days ago

Article

ColBERT-Zero: To Pre-train Or Not To Pre-train ColBERT models?

Feb 19

•

19

upvoted a paper about 1 month ago

Diffusion-Pretrained Dense and Contextual Embeddings

Paper • 2602.11151 • Published Feb 11 • 23

upvoted a collection about 2 months ago

GPT 5 Codex

Distilled models and datasets for GPT 5 Codex • 7 items • Updated Dec 20, 2025 • 5

upvoted an article about 2 months ago

Article

Forge: Scalable Agent RL Framework and Algorithm

Feb 13

•

145

upvoted 2 collections 3 months ago

🧮functiongemma ft mobile-actions

A collection of functiongemma-270m-it models fine-tuned on mobile actions dataset for Spanish, French and Italian • 3 items • Updated Jan 5 • 3

JustRL

2 items • Updated Nov 1, 2025 • 5

upvoted an article 3 months ago

Article

Aligning to What? Rethinking Agent Generalization in MiniMax M2

Oct 30, 2025

•

43

upvoted an article 4 months ago

Article

Encoding the World's Medical Knowledge into 970K

Dec 22, 2025

•

15