Loading activity…
@divy93t
co-lead for science and reasoning capabilities for Gemini @GoogleDeepMind, advancing human-centered llms. Ph.D @CityStGeorges . Personal views.
AI
A post about EdisonSci's long-horizon bioinformatics benchmark says recent model releases dramatically pushed its Pareto frontier.
AI
Glimpse’s team claims competitive frontier performance compared with Astra on key public visual reasoning benchmarks.
AI
The post invites scientists to sign up and says they can turn off telemetry if concerned about data retention.
AI
The poster seeks scientists in life sciences, physical sciences or math who want hands-on data and evaluation work.
AI
A user recommends building an eval instead of settling for a superficial understanding of model performance.
AI
A team member jokes that they can now justify saying “sorry for the late response” to friends and family—for unanswered texts and emails as well as the model release.
AI
Google announces a 1 million-token output limit, up from 64,000, for its new model. Initial access goes to trusted cyber defenders through the Fairwind Program.
AI
Posts say 25 Fields Medal winners signed a declaration warning of a mismatch between AI companies and mathematics. Supporters stress understanding and scientific communities; critics say AI proofs can still yield insights.
AI
Creators share benchmark tables showing Gemini 3.8 Flash results.
AI
Available today for Pro, Max, and Enterprise subscribers on macOS 15+.
AI
Google researchers share early results extending the Co-Scientist tool with Gemini for real-world use.
AI
Investor posted image of April 2023 experiment called Window that handled local and cloud models.
AI
Four former Google researchers found a public benefit corporation to automate machine learning and science.
