Tag: Benchmarks

1. LLMs are shaky judges of whether a research idea is new (huggingface.co)
   AI · 4 days ago · October 30, 2026
2. Nathan Lambert on what a leading AI benchmark can and cannot tell you (interconnects.ai)
   AI · 5 days ago · October 29, 2026
3. AI models can now beat Slay the Spire, but only the easy mode (epoch.ai)
   AI · 1 week ago · October 23, 2026
4. Most working hours fall outside what AI benchmarks measure (mercor.com)
   AI · 2 weeks ago · October 17, 2026
5. ThinkingBox grades AI agents on the database, not the transcript (huggingface.co)
   AI · 3 weeks ago · October 9, 2026
6. AI models jumped from 28% to 80% at spotting IKEA assembly mistakes (epoch.ai)
   AI · 1 month ago · September 24, 2026
7. GPT-6 Astra leads overall but trails Claude on coding (epoch.ai)
   AI · 1 month ago · September 21, 2026
8. A math benchmark of unsolved Erdős problems that models can't fake (epoch.ai)
   AI · 1 month ago · September 14, 2026
9. Cognition's SWE-2 chases cost per task, not the top score (cognition.com)
   AI · 1 month ago · September 11, 2026
10. DeepMind ran a frontier model through a double-blind benchmark (deepmind.google)
    AI · 2 months ago · August 29, 2026
11. Nine questions Epoch wants AI benchmarks to answer (epoch.ai)
    AI · 2 months ago · August 27, 2026
12. Agents still fail most enterprise Java migrations (huggingface.co)
    AI · 4 months ago · July 3, 2026

← all tags