109 56 76

Hugo Laurençon

HugoLaurencon

HugoLaurencon

AI & ML interests

None yet

Recent Activity

upvoted a paper 2 days ago

Qwen2.5 Technical Report

upvoted a paper 2 days ago

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

new activity 2 days ago

HuggingFaceM4/idefics2-8b:Seems like the user prompt is ignored

View all activity

Articles

Docmatix - a huge dataset for Document Visual Question Answering

Jul 18

• 71

Introducing Idefics2: A Powerful 8B Vision-Language Model for the community

Apr 15

• 170

Unlocking the conversion of Web Screenshots into HTML Code with the WebSight Dataset

Mar 15

• 7

Introducing IDEFICS: An Open Reproduction of State-of-the-art Visual Language Model

Aug 22, 2023

• 28

Putting ethical principles at the core of research lifecycle

May 19, 2022

Organizations

Posts 5

Post

2898

We release Idefics2-chatty, the chatbot-optimized version of Idefics2: HuggingFaceM4/idefics2-8b-chatty

Idefics2-chatty is better at following instructions and following Chain-of-Thoughts reasoning.

Moreover, we also release a paper, containing a lot of findings on how to build an efficient and performant Vision-Language Model: What matters when building vision-language models? (2405.02246)

How are you going to use the model, or what data are you going to fine-tune it on?

Post

2514

Idefics2 is trained mostly on OBELICS, our open interleaved image-text document dataset.

Training on interleaved data is crucial to reaching high performance on VQA tasks, taking an arbitrary number of images as input, and doing in-context learning.

Dataset: HuggingFaceM4/OBELICS
Nomic visualization: https://atlas.nomic.ai/map/f2fba2aa-3647-4f49-a0f3-9347daeee499/ee4a84bd-f125-4bcc-a683-1b4e231cb10f
Link to OBELICS thread: https://twitter.com/HugoLaurencon/status/1694005892839006301

View all posts