“Compact Giant” Mistral 7B Outperforms Llama 2 13B and Llama 34B

1 October 2023
Mistral 7B vs Llama 2

“Compact Giant” Mistral 7B Outperforms Llama 2 13B and Llama 34B

The Mistral AI team has unveiled the remarkable Mistral 7B – an open-source language model with a staggering 7.3 billion parameters, surpassing the significantly larger Llama 2 13B model in…

MIT Releases Free Lecture Course on TinyML & Efficient DL Computing on Youtube

29 September 2023
TinyML & Efficient DL Computing

MIT Releases Free Lecture Course on TinyML & Efficient DL Computing on Youtube

In recent years, large language and diffusion models have showcased impressive results. However, their demands on computational resources and memory consumption pose significant challenges for researchers and developers. The TinyML…

Google Apps Integration Elevates Bard Chatbot’s Capabilities

19 September 2023
bard_with_google_services

Google Apps Integration Elevates Bard Chatbot’s Capabilities

Google has rolled out an update for the Bard chatbot, introducing seamless integration with various Google apps, such as Gmail, Docs, Sheets, Maps, and YouTube. This integration propels Bard ahead…

Persimmon-8B: An Open Model with a 16k Token Context, Running on a Single GPU

11 September 2023
persimmon-8b-llm

Persimmon-8B: An Open Model with a 16k Token Context, Running on a Single GPU

Researchers from Adept have introduced the open-source language model Persimmon-8B with a 16k token context, which is four times larger than the most compact Llama 2 and text-davinci-002 used in…

Arthur Bench: Framework for Evaluating Language Models

20 August 2023
arthur bench

Arthur Bench: Framework for Evaluating Language Models

American startup Arthur has released an open-source framework called Bench for evaluating and comparing the performance of large language models. This tool enables users to select the most suitable language…

ReLoRA: Method for Enhancing Performance in Training Large Language Models

16 August 2023
relora method

ReLoRA: Method for Enhancing Performance in Training Large Language Models

ReLoRA is a technique for training large transformer-based language models using low-rank matrices, aimed at boosting training efficiency. The effectiveness of this method increases with the scale of the models.…

Wix AI: Building Websites Using Chatbot Technology

23 July 2023
wix ai

Wix AI: Building Websites Using Chatbot Technology

Wix, the website creation service, has announced the launch of its chatbot, Wix AI, which allows users to create and modify websites using natural language queries. Additionally, the tool will…

Llama 2 and Llama-2-Chat: A New Generation of Open Source Language Models

19 July 2023
Llama 2 update

Llama 2 and Llama-2-Chat: A New Generation of Open Source Language Models

The new generation of Llama models comprises three large language models, namely Llama 2 with 7, 13, and 70 billion parameters, along with the fine-tuned conversational models Llama-2-Chat 7B, 34B,…

Google Bard Update: Image Processing and New Language Support

16 July 2023
google bard

Google Bard Update: Image Processing and New Language Support

Google Bard has undergone an update, expanding its functionality to 46 languages across more than 200 countries, including countries in Europe and Brazil. The latest features include image processing, dialog…

“Deepdub Go” Empowers Content Creators with AI for Video Dubbing

9 July 2023
ai for video dubbing - neural network based service

“Deepdub Go” Empowers Content Creators with AI for Video Dubbing

Israeli startup Deepdub has unveiled its groundbreaking service, Deepdub Go, which utilizes AI for dubbing to automatically dub videos in 65 languages. This innovative platform targets game development studios, advertising…

AI.XYZ: Personal AI Assistant for Personal and Work Tasks

2 July 2023
персональный ИИ ассистент

AI.XYZ: Personal AI Assistant for Personal and Work Tasks

The AI Foundation research lab has launched AI.XYZ, a platform for creating personal AI assistants. The company claims that AI.XYZ is the world’s first platform for managing life using AI,…

Microsoft’s Phi-1 Model with 1.3B Parameters Achieves SotA in Code Generation

30 June 2023
code generation phi-1 model

Microsoft’s Phi-1 Model with 1.3B Parameters Achieves SotA in Code Generation

Researchers at Microsoft Research have introduced Phi-1, a language model for code generation with just 1.3 billion parameters. This model has achieved a state-of-the-art level of code generation using a…

AudioPaLM: Google’s Multimodal Model for Voice Translation

29 June 2023
audiopalm google

AudioPaLM: Google’s Multimodal Model for Voice Translation

Google has introduced AudioPaLM, a large language model for speech processing and generation that combines two Google language models, PaLM-2 and AudioLM, into a multimodal architecture. The model can recognize…

Inflection-1: A Powerful Language Model Surpassing GPT-3.5 in Logical Problem Solving

26 June 2023
Inflection-1 model by Insflection

Inflection-1: A Powerful Language Model Surpassing GPT-3.5 in Logical Problem Solving

Inflection, a startup, officially introduced Inflection-1, a large language model powering the chatbot Pi. Comparable to GPT-3.5, Inflection-1’s size and capabilities match those of ChatGPT. Training took place on “thousands”…

PandasAI: Data Analysis with Language Models

25 June 2023
PandasAI framework

PandasAI: Data Analysis with Language Models

PandasAI is a library that allows performing basic data analysis through natural language queries. Users can specify one or multiple dataframes and a text query, and receive the output in…

LIMA: Pretraining Method on 1000 Examples Achieved GPT4-Level Accuracy

31 May 2023
LIMA LLAMA

LIMA: Pretraining Method on 1000 Examples Achieved GPT4-Level Accuracy

Language models typically undergo two stages of training: unsupervised pretraining and fine-tuning to specific tasks and user preferences. The novel LIMA method (Less Is More for Alignment) challenges the traditional…

IBM has increased the quality of speech recognition by 57% in the Watson Speech to Text service

29 April 2021

IBM has increased the quality of speech recognition by 57% in the Watson Speech to Text service

The improved neural network training strategy has allowed IBM to significantly increase the efficiency of the speech-to-text tool. The service works with eight languages and provides a record high speed…

TextFlint: a library for analyzing NLP models robustness

8 April 2021

TextFlint: a library for analyzing NLP models robustness

TextFlint is a multilingual, multitasking platform for analyzing NLP models stability. Open source available for English and Chinese, other languages ​​are being developed. Included text processing tools: general and specific…

CSTR neural network recognizes text in scene images

1 March 2021

CSTR neural network recognizes text in scene images

CSTR is a convolutional neural network that recognizes text in scene images. The previous work considers the problem of text recognition on a scene image as a segmentation and seq2seq…

TransGAN: two Transformer models as one GAN

26 February 2021

TransGAN: two Transformer models as one GAN

TransGAN is a GAN model in which the generator and discriminator are composed of two Transformer architectures. GAN architecture traditionally uses convolutions. In TransGAN, convolutions are replaced with Transformer. The…

Twitter Opens Tweet Archive for Scientific Researchers

20 February 2021

Twitter Opens Tweet Archive for Scientific Researchers

Twitter has opened an archive of tweets for scientific researchers. This way the IT-company supports research on online discourse and trends on the platform. More data and access to them…