justreadthis.ai
ArchivePeopleAboutSubscribe
justreadthis.ai

A workforce of speciality AI agents read and score hundreds of AI articles every day from top AI newsletters

Don't read every newsletter... just read this.

Get the daily briefing

© 2026 justreadthis.ai

AboutArchivePrivacyTerms
← Back to People
RP
Researcher

Razvan Pascanu

Research Scientist at Google DeepMind

LinkedIn

Razvan Pascanu is a research scientist at Google DeepMind affiliated with Université de Montréal's Department of Computer Science and Operations Research.

1
Mentions
1
Articles
1
Outlets

Associated Topics

Coverage Patterns

How media typically covers Razvan Pascanu

Article Types

  • research1

Media Angles

  • technical1

Narrative Framing

  • cautionary1

Associated AI Models

Transformer1

Articles

Most recent first

Mentioned In

Referenced in coverage

Perplexity Cannot Always Tell Right From Wrong
arXivresearchcautiousFeb 4, 2026

Perplexity is an unsuitable metric for model selection because models with low perplexity can make incorrect predictions, and perplexity does not always favor more accurate models over less accurate ones.

“Co-author of arXiv paper demonstrating that perplexity will not always select for more accurate models despite being widely used.”

AI ResearchFoundation ModelsAI SafetyNLP