Relevant information

about the life in AIRI

Media

Blog

Events

Search results

43 posts

Realistic click simulation will help to improve interactive segmentation models

Realistic click simulation will help to improve interactive segmentation models

Realistic click simulation will help to improve interactive segmentation models

A new benchmark will test the ability of large language models to search for information in long contexts

A new benchmark will test the ability of large language models to search for information in long contexts

The sparse parameter space helps to make transformers more compact

The sparse parameter space helps to make transformers more compact

Hackathon SafeSpeak-2024. Deepfake detection for secure voice communications

Hackathon SafeSpeak-2024. Deepfake detection for secure voice communications

Hackathon SafeSpeak-2024. Deepfake detection for secure voice communications

Large language models proved capable of populating knowledge graphs

Large language models proved capable of populating knowledge graphs

Researchers trained diffusion models to edit images more accurately and faster

Researchers trained diffusion models to edit images more accurately and faster

Researchers improved a reinforcement learning-based model for drug discovery

Researchers improved a reinforcement learning-based model for drug discovery

Neural potential field helped robots to avoid obstacles better

Neural potential field helped robots to avoid obstacles better

Afrikaans and Slovene were found to be the best donors for cross-lingual transfer between high-resource and low-resource languages

Afrikaans and Slovene were found to be the best donors for cross-lingual transfer between high-resource and low-resource languages

Using the general cost functional made the optimal transport with class preservation more accurate

Using the general cost functional made the optimal transport with class preservation more accurate

The new framework required 50 times less data to predict the optimal geometry of molecules using neural networks

The new framework required 50 times less data to predict the optimal geometry of molecules using neural networks

The new framework required 50 times less data to predict the optimal geometry of molecules using neural networks

Language models for chemical tasks turned out to be vulnerable to changes in molecule representations

Language models for chemical tasks turned out to be vulnerable to changes in molecule representations

Gaussian mixtures helped to improve the learning of object-centric scene representation

Gaussian mixtures helped to improve the learning of object-centric scene representation

Energy-based models made Schrödinger Bridge lighter

Energy-based models made Schrödinger Bridge lighter

Recurrent memory removed restrictions on the length of input sequences for transformers

Recurrent memory removed restrictions on the length of input sequences for transformers