-
29
Companionship Leaderboard
π₯AI companionship leaderboard based on the INTIMA benchmark
-
INTIMA: A Benchmark for Human-AI Companionship Behavior
Paper β’ 2508.09998 β’ Published β’ 6 -
Training language models to be warm and empathetic makes them less reliable and more sycophantic
Paper β’ 2507.21919 β’ Published β’ 1
Lynne Devoe
Roxbit
Β·
AI & ML interests
None yet
Recent Activity
upvoted
a
collection
about 9 hours ago
VibeVoice
updated
a Space
4 days ago
Roxbit/wb-kristopher-s-pimp
published
a Space
4 days ago
Roxbit/wb-kristopher-s-pimp
Organizations
None yet