46 38 180

Daniel Vila

dvilasuero

https://argilla.io

AI & ML interests

RLHF, RLAIF, DPO, data, data, data

Articles

Ethics and Society Newsletter #6: Building Better AI: The Importance of Data Quality

Jun 24

• 30

🦙⚗️ Using Llama3 and distilabel to build fine-tuning datasets

Jun 4

• 67

Data is better together

Mar 4

• 6

Organizations

dvilasuero's activity

upvoted a paper 3 days ago

V-STaR: Training Verifiers for Self-Taught Reasoners

Paper • 2402.06457 • Published Feb 9 • 8

upvoted an article 6 days ago

Article

Preference Optimization for Vision Language Models

Jul 10

• 36

upvoted an article 8 days ago

Article

Fine-tuning a token classification model for legal data using Argilla and AutoTrain

•

12 days ago

• 11

upvoted a paper 17 days ago

The Future of Open Human Feedback

Paper • 2408.16961 • Published Aug 15 • 19

upvoted a paper 20 days ago

Direct Language Model Alignment from Online AI Feedback

Paper • 2402.04792 • Published Feb 7 • 27

upvoted an article 24 days ago

Article

Llama-3.1-Storm-8B: Improved SLM with Self-Curation + Model Merging

•

Aug 19

• 72

upvoted a paper 24 days ago

Building and better understanding vision-language models: insights and future directions

Paper • 2408.12637 • Published 28 days ago • 109

upvoted an article about 2 months ago

Article

Llama 3.1 - 405B, 70B & 8B with multilinguality and long context

Jul 23

• 193

upvoted a collection about 2 months ago

Llama 3.1 GPTQ, AWQ, and BNB Quants

Collection

Optimised Quants for high-throughput deployments! Compatible with Transformers, TGI & VLLM 🤗 • 9 items • Updated Jul 24 • 47

upvoted an article 2 months ago

Article

Announcing Finance Commons and the Bad Data Toolbox: Pioneering Open Data and Advanced Document Processing

•

Jul 19

• 17

upvoted a paper 2 months ago

DataDream: Few-shot Guided Dataset Generation

Paper • 2407.10910 • Published Jul 15 • 7

upvoted 2 articles 2 months ago

Article

SmolLM - blazingly fast and remarkably powerful

Jul 16

• 242

Article

How we leveraged distilabel to create an Argilla 2.0 Chatbot

Jul 16

• 30

upvoted an article 3 months ago

Article

Data Is Better Together: A Look Back and Forward

Jun 20

• 17

upvoted a paper 3 months ago

Show, Don't Tell: Aligning Language Models with Demonstrated Feedback

Paper • 2406.00888 • Published Jun 2 • 30

upvoted a paper 4 months ago

Are You Sure? Rank Them Again: Repeated Ranking For Better Preference Datasets

Paper • 2405.18952 • Published May 29 • 10

upvoted 2 articles 4 months ago

Article

⚗️ 🔥 Building High-Quality Datasets with distilabel and Prometheus 2

•

Jun 3

• 26

Article

Training and Finetuning Embedding Models with Sentence Transformers v3

May 28

• 146

upvoted a paper 4 months ago

RLHF Workflow: From Reward Modeling to Online RLHF

Paper • 2405.07863 • Published May 13 • 67

upvoted 4 articles 5 months ago

Article

Can we create pedagogically valuable multi-turn synthetic datasets from Cosmopedia?

•

May 7

• 7

Article

Data is better together

Mar 4

• 6

Article

🧑‍⚖️ "Replacing Judges with Juries" using distilabel

•

May 3

• 17

Article

⚗️ 🧑🏼‍🌾 Let's grow some Domain Specific Datasets together

•

Apr 29

• 28

upvoted a collection 5 months ago

Zephyr ORPO

Collection

Models and datasets to align LLMs with Odds Ratio Preference Optimisation (ORPO). Recipes here: https://github.com/huggingface/alignment-handbook • 3 items • Updated Apr 12 • 16

upvoted 2 collections 6 months ago

Creación de corpus en comunidad

Collection

Colección de esfuerzos colaborativos para crear corpus en español de calidad. Toda persona hispanohablante puede contribuir :) • 7 items • Updated Jul 17 • 6

DIBT Prompt collective SPIN

Collection

This collection contains resources related to the replication of SPIN with the dibt prompt collective dataset • 8 items • Updated Jul 30 • 7

upvoted 4 collections 8 months ago

Apple MLX-compatible 7B LLMs on the 🤗 Hub

Collection

This collection contains the model weights for 7B LLMs for Apple's MLX framework. Find more information at https://github.com/ml-explore/mlx • 8 items • Updated 17 days ago • 9

Datasets built with ⚗️ distilabel

Collection

This collection contains some datasets generated and/or labelled using https://github.com/argilla-io/distilabel • 7 items • Updated Aug 6 • 9

Preference Datasets for DPO

Collection

This collection contains a list of curated preference datasets for DPO fine-tuning for intent alignment of LLMs • 7 items • Updated Jul 30 • 28

Comparing DPO with IPO and KTO

Collection

A collection of chat models to explore the differences between three alignment techniques: DPO, IPO, and KTO. • 56 items • Updated Jan 9 • 31

upvoted a paper 9 months ago

Principled Instructions Are All You Need for Questioning LLaMA-1/2, GPT-3.5/4

Paper • 2312.16171 • Published Dec 26, 2023 • 34

upvoted a collection 9 months ago

Awesome feedback datasets

Collection

A curated list of datasets with human or AI feedback. Useful for training reward models or applying techniques like DPO. • 19 items • Updated Apr 12 • 64

upvoted a collection 10 months ago

Notus 7B v1

Collection

Notus 7B v1 models (DPO fine-tune of Zephyr SFT) and datasets used. More information at https://github.com/argilla-io/notus • 11 items • Updated Jul 30 • 17

upvoted a paper 10 months ago

Prometheus: Inducing Fine-grained Evaluation Capability in Language Models

Paper • 2310.08491 • Published Oct 12, 2023 • 53

upvoted a collection 10 months ago

OpenChat

Collection

OpenChat: Advancing Open-source Language Models with Mixed-Quality Data • 7 items • Updated Jul 31 • 33

upvoted a paper 12 months ago

Adapting Large Language Models via Reading Comprehension

Paper • 2309.09530 • Published Sep 18, 2023 • 75

Daniel Vila

AI & ML interests

Articles

🔥 Argilla 2.0: the data-centric tool for AI makers 🤗

Llama 3.1 - 405B, 70B & 8B with multilinguality and long context

How we leveraged distilabel to create an Argilla 2.0 Chatbot

Ethics and Society Newsletter #6: Building Better AI: The Importance of Data Quality

🦙⚗️ Using Llama3 and distilabel to build fine-tuning datasets

Data is better together

Organizations

dvilasuero's activity

Preference Optimization for Vision Language Models

Fine-tuning a token classification model for legal data using Argilla and AutoTrain

Llama-3.1-Storm-8B: Improved SLM with Self-Curation + Model Merging

Llama 3.1 - 405B, 70B & 8B with multilinguality and long context

Announcing Finance Commons and the Bad Data Toolbox: Pioneering Open Data and Advanced Document Processing

SmolLM - blazingly fast and remarkably powerful

How we leveraged distilabel to create an Argilla 2.0 Chatbot

Data Is Better Together: A Look Back and Forward

⚗️ 🔥 Building High-Quality Datasets with distilabel and Prometheus 2

Training and Finetuning Embedding Models with Sentence Transformers v3

Can we create pedagogically valuable multi-turn synthetic datasets from Cosmopedia?

Data is better together

🧑‍⚖️ "Replacing Judges with Juries" using distilabel

⚗️ 🧑🏼‍🌾 Let's grow some Domain Specific Datasets together