Pedro Cuenca's picture

Pedro Cuenca

pcuenq

·

AI & ML interests

None yet

Recent Activity

reacted to m-ric's post with 🚀 about 7 hours ago

We now have a Deep Research for academia: SurveyX automatically writes academic surveys nearly indistinguishable from human-written ones 🔥 Researchers from Beijing and Shanghai just published the first application of a deep research system to academia: their algorithm, given a question, can give you a survey of all papers on the subject. To make a research survey, you generally follow two steps, preparation (collect and organize papers) and writing (outline creation, writing, polishing). Researchers followed the same two steps and automated them. 🎯 For the preparation part, a key part is find all the important references on the given subject. Researchers first cast a wide net of all relevant papers. But then finding the really important ones is like distilling knowledge from a haystack of information. To solve this challenge, they built an “AttributeTree” object that structures key information from citations. Ablating these AttributeTrees significantly decreased structure and synthesis scores, so they were really useful! 📝 For the writing part, key was to get a synthesis that's both short and true. This is not easy to get with LLMs! So they used methods like LLM-based deduplication to shorten the too verbose listings made by LLMs, and RAG to grab original quotes instead of made-up ones. As a result, their system outperforms previous approaches by far! As assessed by LLM-judges, the quality score os SurveyX even approaches this of human experts, with 4.59/5 vs 4.75/5 🏆 I advise you to read the paper, it's a great overview of the kind of assistants that we'll get in the short future! 👉 https://huggingface.co/papers/2502.14776 Their website shows examples of generated surveys 👉 http://www.surveyx.cn/

new activity about 8 hours ago

agents-course/course-images:Rename en/unit2/smolagents/afred-party.jpg to en/unit2/smolagents/alfred-party.jpg

updated a dataset about 8 hours ago

agents-course/course-images

View all activity

Organizations

pcuenq's activity

upvoted an article 1 day ago

Article

Remote VAEs for decoding with HF endpoints 🤗

2 days ago

• 25

upvoted a collection 4 days ago

SigLIP2

36 items • Updated 4 days ago • 46

upvoted an article 4 days ago

Article

SigLIP 2: A better multilingual vision language encoder

5 days ago

• 90

upvoted an article 5 days ago

Article

SmolVLM2: Bringing Video Understanding to Every Device

6 days ago

• 162

upvoted a collection 6 days ago

PaliGemma 2 Mix

13 items • Updated 6 days ago • 59

upvoted an article 6 days ago

Article

PaliGemma 2 Mix - New Instruction Vision Language Models by Google

7 days ago

• 58

upvoted an article 7 days ago

Article

Introducing Three New Serverless Inference Providers: Hyperbolic, Nebius AI Studio, and Novita 🔥

8 days ago

• 89

upvoted a paper 8 days ago

BIOMEDICA: An Open Biomedical Image-Caption Archive, Dataset, and Vision-Language Models Derived from Scientific Literature

Paper • 2501.07171 • Published Jan 13 • 50

upvoted a paper 11 days ago

Scaling Pre-training to One Hundred Billion Data for Vision Language Models

Paper • 2502.07617 • Published 14 days ago • 28

upvoted a collection 12 days ago

Qwen2-VL

Vision-language model series based on Qwen2 • 16 items • Updated Dec 6, 2024 • 207

upvoted an article 14 days ago

Article

Object Detection Leaderboard

Sep 18, 2023

• 9

upvoted a collection 15 days ago

Hibiki fr-en

Hibiki is a model for streaming speech translation , which can run on device! See https://github.com/kyutai-labs/hibiki. • 5 items • Updated 19 days ago • 50

upvoted a collection 16 days ago

SmolLM2

State-of-the-art compact LLMs for on-device applications: 1.7B, 360M, 135M • 16 items • Updated 5 days ago • 240

upvoted a paper 18 days ago

Stable Flow: Vital Layers for Training-Free Image Editing

Paper • 2411.14430 • Published Nov 21, 2024 • 22

upvoted a collection 18 days ago

Diffusers Guides

Collection of diffusers guides and their respective spaces • 2 items • Updated Oct 9, 2024 • 2

upvoted an article 20 days ago

Article

Open-source DeepResearch – Freeing our search agents

22 days ago

• 1.1k

upvoted 2 collections 21 days ago

DeepSeek R1 (All Versions)

DeepSeek R1 - the most powerful reasoning open-source model - available in GGUF, original & 4-bit formats. Includes Llama & Qwen distilled models. • 29 items • Updated 18 days ago • 199

January 31 Releases 🧤

24 items • Updated 25 days ago • 7

upvoted 2 articles 21 days ago

Article

SmolVLM Grows Smaller – Introducing the 250M & 500M Models!

Jan 23

• 142

Article

FineVideo: behind the scenes

Sep 23, 2024

• 29