84
Zebra Logic Bench
🦓
Display and explore zebra puzzle leaderboard
Zebra Logic Bench
ZebraLogic Bench: Testing the Limits of LLMs in Logical Reasoning
-
allenai/ZebraLogicBench
Viewer • Updated • 4.26k • 237 • 11Note 1k puzzles in grid format and 3k in mcqa format. answers are hidden.
allenai/ZebraLogicBench-private
Viewer • Updated • 4.26k • 34.5k • 8Note answers are included here. to prevent data leakage, plz apply for access.
Faith and Fate: Limits of Transformers on Compositionality
Paper • 2305.18654 • Published • 6Note Our NeurIPS 2023 (spotlight) paper about the data and analysis.