AI Research Tools: TasteVal Measures “Research Taste”, Karotte Targets Reward Hacking and Zephon Fixes Data Order
A new benchmark, TasteVal, says the best AI model now reaches expert-level results on eight AI research tasks with 2.3 times less compute, though its authors list reasons that may overstate progress. Preference Model has open-sourced Karotte, a framework for training environments that are harder to game, and DatologyAI has released Zephon, a data loader that keeps experiments repeatable when the GPU count changes. What each claims, and who measured it.
Just in
Latest
No posts to display
Timeline
Every story in research taste, grouped by month.
October 2026
- 8 Oct 2026: AI Research Tools: TasteVal Measures "Research Taste", Karotte Targets Reward Hacking and Zephon Fixes Data OrderA new benchmark, TasteVal, says the best AI model now reaches expert-level results on eight AI research tasks with 2.3 times less compute, though…
Latest on TSN
Explore the TSN Atlas: Every topic we cover, mapped. Click a bubble to dive in.
