📃 Multilayer Clustered Graph Learning

🗓 Publish year: 2020

🧑‍💻Authors: Mireille El Gheche, Pascal Frossard
🏢 Universities:  Ecole Polytechnique Fed´ erale de Lausanne (EPFL)

📎 Study the paper

✅ https://news.1rj.ru/str/DataScienceT

📃SimTeG: A Frustratingly Simple Approach Improves Textual Graph Learning

🗓 Publish year: 2023

🧑‍💻Authors: Keyu Duan, Qian Liu,Tat-Seng Chua, Shuicheng Yan, Wei Tsang Ooi, Qizhe Xie, Junxian He
🏢Universities: ENational University of Singapore, The Hong Kong University of Science and Technology

📎 Study the paper

💻 Code

✅ https://news.1rj.ru/str/DataScienceT

🪐 YOLO-CIANNA: Neural Astro🪐

😏 CIANNA is a general-purpose deep learning framework for (but not only for) astronomical data analysis. Source Code released 💙

👉 Review: https://t.ly/441XS

👉 Paper: arxiv.org/pdf/2402.05925.pdf

👉 Code: github.com/Deyht/CIANNA

👉 Wiki: github.com/Deyht/CIANNA/wiki

✅ https://news.1rj.ru/str/DataScienceT

📁 A Survey of Graph Neural Networks in Real world: Imbalance, Noise, Privacy and OOD Challenges

🗓 Publish year: 2024

🧑‍💻 Authors: Wei Ju, Siyu Yi, Yifan Wang, Zhiping Xiao, Zhengyang Mao, Hourun Li, Yiyang Gu, Yifang Qin, Nan Yin, Senzhang Wang, Xinwang Liu, Xiao Luo, Philip S. Yu, Ming Zhang

📎 Study the paper

Taming Stable Diffusion for Text to 360° Panorama Image Generation

🖥 Github: https://github.com/chengzhag/panfusion

📕 Paper: https://arxiv.org/abs/2404.07949v1

🔥 Dataset: https://chengzhag.github.io/publication/panfusion/

✅ https://news.1rj.ru/str/DataScienceT

✨ HairFastGAN: Realistic and Robust Hair Transfer with a Fast Encoder-Based Approach

A new model for transferring a hairstyle from a reference image to a source photo for a virtual fitting room.

▪ Paper : https://arxiv.org/abs/2404.01094

▪ Code : https://github.com/AIRI-Institute/HairFastGAN

▪ Colab : https://colab.research.google.com/#fileId=https%3A//huggingface.co/AIRI-Institute/HairFastGAN/blob/main/notebooks/HairFast_inference.ipynb

✅ https://news.1rj.ru/str/DataScienceT

Are you fluent in Python and want to evaluate your skills? 🤔

Do you want to learn Python? 🤗

Are you interested in learning through questions and answers? 😵

Do you want to receive the explanation of the question?💡

🟢 https://news.1rj.ru/str/DataScienceQ 👍

🌟 Not allowed to use ChatGPT - LLM deployments locally

There are situations when life circumstances do not allow using ChatGPT and you have to deploy LLM locally.
What can be used in this case?

1. Proprietary models :
🟡 Anthropic - Currently comparable to or superior to ChatGPT 4.0 on some tasks and has a large context window, making it possible to solve many problems without resorting to RAG and other hybrid methods

🟡 Yandex GPT - functions well in Russian, so if your grandmother is also a major, she will definitely appreciate this option

🟡 GigaChat is a model from Sberbank, it also works well in Russian and see the point above

2. Open models :
🟡 LLama 2 is an original open model from a well-known terrorist organization, on the basis of which over 100,500 different models have already been piled up, for which many thanks to this organization (still no one understands what prompted Mark to make this decision). The quality is not up to ChatGPT 4.

🟡 ruGPT is a pretrain from GigaChat under the MIT license. Sber had a hand here too, thanks to them. Can be used

🟡 Mistral is a model developed by people from Google in France. The quality is not up to ChatGPT 4, but on average it is better than Llama 2.

🟡 Falcon is a model developed with Arab money by Europeans. Overall, Llama 2 is weaker, and the point of using it eludes me.

🟡 Grok from X is supposedly a “based” model from Elon himself. It works so-so so far, give or take at the level of ChatGPT 3.5, but Elon promises to tear everyone to rags and there are reasons to believe him.

Model estimates currently look something like this (pictured)

✅ https://news.1rj.ru/str/DataScienceT

👑 Llama 3 is here, with a brand new tokenizer! 🦙

Llama 3 released

Meta has released the new SOTA Llama 3 in two versions for 8B and 70B parameters.

Context length 8K, support 30 languages.

• HF : https://huggingface.co/spaces/ysharma/Chat_with_Meta_llama3_8b
• Blog : https://ai.meta.com/blog/meta-llama-3/

You can test 🦙 MetaLlama 3 70B and 🦙 Meta Llama 3 8B using the 🔥 free interface: https://llama3.replicate.dev/

✅ https://news.1rj.ru/str/DataScienceT