Blog, Articles, and discussions

Putting RL back in RLHF

By June 12, 2024 • 100

Community Articles

view all

NVIDIA Releases 3 Million Sample Dataset for OCR, Visual Question Answering, and Captioning Tasks

and 4 others •

1 day ago

• 24

<p style="text-align:center;"> Bridging the Gap: Making Robotics Feel Like Machine Learning </p>

•

about 23 hours ago

• 9

OpenAI just dropped two massive open-weight models — but how do we separate the reality from the hype?

and 2 others •

4 days ago

• 9

Uncensor any LLM with abliteration

•

Jun 13, 2024

• 647

DeepSeek-R1 Dissection: Understanding PPO & GRPO Without Any Prior Reinforcement Learning Knowledge

•

Feb 7

• 203

Building Enterprise-Ready Text Classifiers in Minutes with Adaptive Learning

•

4 days ago

• 8

From GRPO to DAPO and GSPO: What, Why, and How

•

4 days ago

• 8

What Open-Source Developers Need to Know about the EU AI Act's Rules for GPAI Models

and 5 others •

9 days ago

• 23

The Missing Semester of AI for Organizations #1: LLM Security

•

7 days ago

• 7

Luth: Efficient French Specialization for Small Language Models

and 1 other •

2 days ago

• 6

LLM agent experiment with a purpose-built RPG and tool calls. (Work in progress)

•

8 days ago

• 7

Introducing : 🤏🏻🏭SmolFactory

•

3 days ago

• 5

Towards Open Evolutionary Agents

and 1 other •

9 days ago

• 13

What I Learned Upscaling a Long-distance Midjourney Photo w/ Stable Diffusion PLUS unboxing Qwen Image & Wan 2.2

•

4 days ago

• 5

How I Built 7 Custom Gradio Components in Just 12 Days!

•

about 21 hours ago

• 5

Jack of All Trades, Master of Some, a Multi-Purpose Transformer Agent

By April 22, 2024 • 81

Constitutional AI with Open LLMs

By February 1, 2024 • 15

Preference Tuning LLMs with Direct Preference Optimization Methods

By January 18, 2024 • 70

The N Implementation Details of RLHF with PPO

By October 24, 2023 • 65

Finetune Stable Diffusion Models with DDPO via TRL

By September 29, 2023 guest • 16

Fine-tune Llama 2 with DPO

By August 8, 2023 • 60

StackLLaMA: A hands-on guide to train LLaMA with RLHF

By April 5, 2023 • 42

Fine-tuning 20B LLMs with RLHF on a 24GB consumer GPU

By March 9, 2023 • 60

Introducing ⚔️ AI vs. AI ⚔️ a deep reinforcement learning multi-agents competition system

By February 7, 2023 • 2

Illustrating Reinforcement Learning from Human Feedback (RLHF)

By December 9, 2022 • 316

Train your first Decision Transformer

By September 8, 2022 • 13

Proximal Policy Optimization (PPO)

By August 5, 2022 • 53

Advantage Actor Critic (A2C)

By July 22, 2022 • 5

Policy Gradient with PyTorch

By June 30, 2022

Community Articles

NVIDIA Releases 3 Million Sample Dataset for OCR, Visual Question Answering, and Captioning Tasks

and 4 others •

1 day ago

• 24

AWorld Multi-Agent System Hits #1 on GAIA Leaderboard

•

6 days ago

• 22

RynnVLA-001: Using Human Demonstrations to Improve Robot Manipulation

and 9 others •

2 days ago

• 19

The GPT-OSS models are here… and they’re energy-efficient!

•

6 days ago

• 15

Code a simple RAG from scratch

•

Oct 29, 2024

• 150

ChatML vs Harmony: Understanding the new Format from OpenAI 🔍

•

4 days ago

• 11

<p style="text-align:center;"> Bridging the Gap: Making Robotics Feel Like Machine Learning </p>

•

about 23 hours ago

• 9

OpenAI just dropped two massive open-weight models — but how do we separate the reality from the hype?

and 2 others •

4 days ago

• 9

Uncensor any LLM with abliteration

•

Jun 13, 2024

• 647

DeepSeek-R1 Dissection: Understanding PPO & GRPO Without Any Prior Reinforcement Learning Knowledge

•

Feb 7

• 203

Building Enterprise-Ready Text Classifiers in Minutes with Adaptive Learning

•

4 days ago

• 8

From GRPO to DAPO and GSPO: What, Why, and How

•

4 days ago

• 8

What Open-Source Developers Need to Know about the EU AI Act's Rules for GPAI Models

and 5 others •

9 days ago

• 23

The Missing Semester of AI for Organizations #1: LLM Security

•

7 days ago

• 7

Luth: Efficient French Specialization for Small Language Models

and 1 other •

2 days ago

• 6

LLM agent experiment with a purpose-built RPG and tool calls. (Work in progress)

•

8 days ago

• 7

Introducing : 🤏🏻🏭SmolFactory

•

3 days ago

• 5

Towards Open Evolutionary Agents

and 1 other •

9 days ago

• 13

What I Learned Upscaling a Long-distance Midjourney Photo w/ Stable Diffusion PLUS unboxing Qwen Image & Wan 2.2

•

4 days ago

• 5

How I Built 7 Custom Gradio Components in Just 12 Days!

•

about 21 hours ago

• 5

View all

Blog, Articles, and discussions

Putting RL back in RLHF

NVIDIA Releases 3 Million Sample Dataset for OCR, Visual Question Answering, and Captioning Tasks

AWorld Multi-Agent System Hits #1 on GAIA Leaderboard

RynnVLA-001: Using Human Demonstrations to Improve Robot Manipulation

The GPT-OSS models are here… and they’re energy-efficient!

Code a simple RAG from scratch

ChatML vs Harmony: Understanding the new Format from OpenAI 🔍

<p style="text-align:center;"> Bridging the Gap: Making Robotics Feel Like Machine Learning </p>

OpenAI just dropped two massive open-weight models — *but how do we separate the reality from the hype?*

Uncensor any LLM with abliteration

DeepSeek-R1 Dissection: Understanding PPO & GRPO Without Any Prior Reinforcement Learning Knowledge

Building Enterprise-Ready Text Classifiers in Minutes with Adaptive Learning

From GRPO to DAPO and GSPO: What, Why, and How

What Open-Source Developers Need to Know about the EU AI Act's Rules for GPAI Models

The Missing Semester of AI for Organizations #1: LLM Security

Luth: Efficient French Specialization for Small Language Models

LLM agent experiment with a purpose-built RPG and tool calls. (Work in progress)

Introducing : 🤏🏻🏭SmolFactory

Towards Open Evolutionary Agents

What I Learned Upscaling a Long-distance Midjourney Photo w/ Stable Diffusion PLUS unboxing Qwen Image & Wan 2.2

How I Built 7 Custom Gradio Components in Just 12 Days!

Jack of All Trades, Master of Some, a Multi-Purpose Transformer Agent

Constitutional AI with Open LLMs

Preference Tuning LLMs with Direct Preference Optimization Methods

The N Implementation Details of RLHF with PPO

Finetune Stable Diffusion Models with DDPO via TRL

Fine-tune Llama 2 with DPO

StackLLaMA: A hands-on guide to train LLaMA with RLHF

Fine-tuning 20B LLMs with RLHF on a 24GB consumer GPU

Introducing ⚔️ AI vs. AI ⚔️ a deep reinforcement learning multi-agents competition system

Illustrating Reinforcement Learning from Human Feedback (RLHF)

Train your first Decision Transformer

Proximal Policy Optimization (PPO)

Advantage Actor Critic (A2C)

Policy Gradient with PyTorch

NVIDIA Releases 3 Million Sample Dataset for OCR, Visual Question Answering, and Captioning Tasks

AWorld Multi-Agent System Hits #1 on GAIA Leaderboard

RynnVLA-001: Using Human Demonstrations to Improve Robot Manipulation

The GPT-OSS models are here… and they’re energy-efficient!

Code a simple RAG from scratch

ChatML vs Harmony: Understanding the new Format from OpenAI 🔍

<p style="text-align:center;"> Bridging the Gap: Making Robotics Feel Like Machine Learning </p>

OpenAI just dropped two massive open-weight models — *but how do we separate the reality from the hype?*

Uncensor any LLM with abliteration

DeepSeek-R1 Dissection: Understanding PPO & GRPO Without Any Prior Reinforcement Learning Knowledge

Building Enterprise-Ready Text Classifiers in Minutes with Adaptive Learning

From GRPO to DAPO and GSPO: What, Why, and How

What Open-Source Developers Need to Know about the EU AI Act's Rules for GPAI Models

The Missing Semester of AI for Organizations #1: LLM Security

Luth: Efficient French Specialization for Small Language Models

LLM agent experiment with a purpose-built RPG and tool calls. (Work in progress)

Introducing : 🤏🏻🏭SmolFactory

Towards Open Evolutionary Agents

What I Learned Upscaling a Long-distance Midjourney Photo w/ Stable Diffusion PLUS unboxing Qwen Image & Wan 2.2

How I Built 7 Custom Gradio Components in Just 12 Days!

OpenAI just dropped two massive open-weight models — but how do we separate the reality from the hype?

OpenAI just dropped two massive open-weight models — but how do we separate the reality from the hype?