General365 benchmark reveals that even leading LLMs achieve only 62.8% accuracy on general reasoning tasks limited to K-12 knowledge, showing domain-dependent capabilities unlike their strong math and physics performance.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
General365: Benchmarking General Reasoning in Large Language Models Across Diverse and Challenging Tasks
General365 benchmark reveals that even leading LLMs achieve only 62.8% accuracy on general reasoning tasks limited to K-12 knowledge, showing domain-dependent capabilities unlike their strong math and physics performance.