#Research
Curated coverage on «Research»: releases, research, benchmarks, open-source.
Google, Microsoft and xAI to let the US review AI models before release
Three of the biggest AI companies have agreed to let the US government review new models before their public release. The government review program is moving from a…
Ten Minutes with AI Reduce Cognitive Abilities: Study Findings
A new study found that even a short 10-minute session with an AI assistant reduces a person’s ability to think independently and solve problems.
Altman testifies: Musk caused major harm to OpenAI's culture
During the trial, OpenAI CEO Sam Altman said Elon Musk caused serious damage to the company's culture by demanding mass cuts in the research division.
AWS showed how to build an offline feature store in SageMaker Unified Studio and Catalog
Diffrax and JAX: a practical guide to ODEs, stochastic simulations, and neural ODE
Cursor questioned public AI benchmarks for code with five charts
Theia Insights raises $8 million to replace outdated industry classification systems
Selectel: 35% of Russian companies increased IT capacity for AI projects over the past year
A Selectel study found that half of Russian companies already use infrastructure for AI, and the main barriers are now tied not to demand, but to economic factors…
Suricata showed how to train ML-based attack detection systems on real traffic
The study showed that Suricata IDS events can be used to label real network traffic and train ML-based IDS, if signatures are carefully selected and noisy labels…
John McCarthy and the myth of “AI”: why CIP is a more accurate term for the technology
MIT report: AI will reshape text-based work gradually rather than eliminate professions all at once
ZDNet: BlueOptima showed that AI for code in production is noticeably weaker than promised
UCL and Stanford researcher explains why AI benchmarks no longer work
MWS AI and SberAI Propose DRAGOn — a Dynamic Benchmark for Evaluating RAG Systems
Researchers from MWS AI, SberAI, and universities described DRAGOn — an approach to dynamic RAG evaluation where the corpus is regularly updated and questions and…
VK presented three student projects from EMS — from a multimodal benchmark to text analysis
VK presented three projects by EMS students: a multimodal benchmark for Russian, a lower-cost attention mechanism, and a system for analyzing emotions and states…