SLR: An Automated Synthesis Framework for Scalable Logical Reasoning
Paper • 2506.15787 • Published • 3
Scalable Logical Reasoning
A collection of scalable logical reasoning tasks
-
-
AIML-TUDA/SLR-Bench
Viewer • Updated • 38.5k • 1.22k • 4 -
AIML-TUDA/SLR-Bench-German
Viewer • Updated • 38.5k • 1.99k • 2 -
AIML-TUDA/SLR-Bench-Spanish
Viewer • Updated • 38.5k • 734 • 1 -
AIML-TUDA/SLR-Bench-French
Viewer • Updated • 38.5k • 608 • 1 -
AIML-TUDA/SLR-Bench-Italian
Viewer • Updated • 38.5k • 177 -
AIML-TUDA/SLR-Bench-Portuguese
Viewer • Updated • 27k • 877 -
AIML-TUDA/Llama-3.1-8B-SLR
8B • Updated • 1 • 1 -
AIML-TUDA/SLR-Bench-Dutch
Viewer • Updated • 38.5k • 990 -
AIML-TUDA/Qwen3-8B-SLR
8B • Updated • 2 • 1 -
VerifiableRewardsForScalableLogicalReasoning
🚀1Evaluate logical rules with a validation program
-
AIML-TUDA/SLR-Homes
Viewer • Updated • 500 • 92 • 1 -
SLR-Bench Leaderboard - Reward Hacking in Reasoning Models
🎯1Reward shortcut behavior in LLMs via IPT