Hippo-RAG: Long-term memory for LLM inspired by neurobiology
Scientists are often inspired by our or animal biological structures: CNN, MLP, Backprop, and for many other studies, similarities can be found in the real world. Recently,…
Introduction to Large Language Models (LLM)
Applications and Benefits of LLM in IT
Neural networks, scammers, and "mammoths": how artificial intelligence is changing cyber fraud
Remember how we were taught not to talk to strangers as children? In 2024, this wisdom has taken on a new meaning: now you can't even be sure that you are talking to…
Does Vision Llama understand impressionists?
Hello everyone, my name is Arseniy, I am a Data Scientist at Raft, and today I will tell you about Visual Language Models.
How I Automated Job Search Using LLM
In light of recent news about layoffs in the IT industry, the issue of job searching is becoming increasingly acute even for people who are currently employed. It has…
A little about Chain-of-Thoughts
Main problems of LLM:
How to evaluate an LLM model
In one of the previous blogs, I introduced the concept of testing large language models. However, testing large language models is a rather complex topic that requires…
The partnership between Microsoft and OpenAI is starting to crack
“The best bromance” in the tech world has faced a harsh reality: OpenAI tried to change the terms of the deal with Microsoft, while the software giant sought ways to…
Evaluation of LLM: metrics, frameworks, and best practices
Jensen Huang in his speech at the "Data+AI" summit said: "Generative AI is everywhere, in every industry. If there is no generative AI in your industry yet, then you…
LLaMator: Red Teaming framework for testing LLM vulnerabilities
Hello, tekkix! In this material, we, the LLaMasters team — 1st year master's students of ITMO AI Talent Hub, present the LLaMator framework, which won the AI Product…
IQ distribution: how a programmer can survive in a world designed for fools?
You have probably often caught yourself thinking that many things in our world are arranged as illogically as possible, and, ideally, should have been arranged completely…
Using ChatGPT for automating code generation in a JS project
In my previous articles on developing Telegram bots, I shared my experience of creating basic functionality and working with commands through arguments. While preparing…
From context to jurisdiction: 7 key parameters when choosing an LLM for your project
A year ago, it seemed that simply choosing GPT-4 was enough to solve all AI problems. Today, the language model market resembles a zoo, with new exotic species appearing…
AI without illusions. Debunking myths
In my podcast, I threatened to read the article GSM-Symbolic: Understanding the Limitations of Mathematical Reasoning in Large Language Models by Apple scientists and…
How to teach LLM to understand video? Overview of approaches
Hello everyone! Today we will talk about the task of understanding video and the evolution of approaches to training multimodal large language models for this task.
How to Win Kaggle Competitions: LLM Analysis of Winning Solutions
About a year ago, a Kaggle enthusiast named Darek Kłeczek conducted an interesting experiment: he collected all available descriptions of winning solutions on Kaggle…
Comparison of five LLM models: ChatGPT-4, GigaChat Pro, GigaChat Lite, YaGPT Pro, and Llama 3 7B
Hello, friends! Today we would like to discuss five popular LLM models that our team had the opportunity to work with: ChatGPT-4, GigaChat Pro, GigaChat Lite, YaGPT Pro,…
Models Gemini 1.5 from Google: main about Gemini 1.5 Pro
One of the main models of the Gemini 1.5 series is the Gemini 1.5 Pro. This is the latest version of large language models from Google, which has attracted significant…
What can gorilla Koko teach us
On July 4, 1971, the 50th gorilla born in captivity was born at the San Francisco Zoo. Since the birth date of this gorilla coincides with the US Independence Day, which…