Pinned Loading
-
-
task-gaming
task-gaming PublicPre-registered study: does the effort cost of the honest path drive task-gaming rates? Pre-commit Hook environment, E in {10,51,258}, 300 counted rollouts.
Python
-
trace-ai-mvp
trace-ai-mvp PublicJudge-reliability MVP: measures where an LLM judge can be trusted to rank human-AI collaboration transcripts, and abstains where it can't. Two-stage evidence-gathering judge + Bradley-Terry ranking…
Python
-
Something went wrong, please refresh the page to try again.
If the problem persists, check the GitHub status page or contact support.
If the problem persists, check the GitHub status page or contact support.

