AI
SageMath拡張LLMエージェントの計算および実験数学における評価
Evaluating SageMath-Augmented LLM Agents for Computational and Experimental Mathematics
arxiv2026年7月9日
日本語要約
本論文は、計算数学および実験数学のタスクのためにSageMathで拡張されたLLMエージェントを評価する。これは、大規模言語モデルと記号計算システム間の相乗効果を探求する。この統合により、AIはより洗練された数学的問題に取り組み、科学的発見を加速できるようになる可能性がある。
English Summary
This paper evaluates LLM agents augmented with SageMath for tasks in computational and experimental mathematics. It explores the synergy between large language models and symbolic computation systems. This integration could empower AI to tackle more sophisticated mathematical problems and accelerate scientific discovery.