A Benchmark for LLM's Understanding of Middle School and High School Science Topics.

arXiv:2609.32020v1 Announce Type: new Abstract: Large language models (LLMs) are increasingly integrated into educational settings, yet educators lack robust, standards-aligned tools to evaluate their effectiveness in K-12 science contexts. Existing benchmarks