Search papers, labs, and topics across Lattice.
This study integrates SageMath with LLMs in a ReAct-style agentic framework to enhance the performance of AI in solving complex mathematical problems. By evaluating this setup against the RealMath benchmark, the authors introduce a multi-step post-processing and validation pipeline that significantly improves the quality of problem sets. The results show an average performance increase of 9.7 percentage points across models, with notable gains for Qwen 3.7-Max and a peak solve rate of 75.2% for GPT-5.5, highlighting the potential of CAS-augmented agents in computational mathematics.
SageMath integration boosts LLM performance in solving advanced mathematical problems by nearly 10 percentage points, narrowing the gap between open and closed models.
Recent advances in AI for Mathematics have focused largely on autoformalization and theorem proving, leaving the role of Computer Algebra Systems (CAS) in agentic LLM workflows underexplored. We propose a ReAct-style agentic setup that combines LLM reasoning with verifiable feedback from SageMath, together with Context7 for the up-to-date documentation. We evaluate this agentic setup across frontier models for solving research-level mathematical problems from the RealMath benchmark in a setting that emulates a computational-mathematics research loop. We also propose a refinement to the RealMath benchmark by introducing a multi-step post-processing procedure and a multi-stage validation pipeline, both of which improve the quality and reliability of the extracted problem set. Our experiments reveal substantial performance gains from SageMath access across all evaluated models on +9.7~pp on average, the gains range from 1.5~pp to 27.8~pp and narrow the gap between open-weight and closed models. Qwen~3.7-Max benefits from SageMath the most, while GPT-5.5 achieves the highest solve rate of $75.2\%$ and the lowest token usage among tool-enabled configurations. Our findings suggest that CAS-augmented agents represent a promising direction for assisting mathematicians in computational exploration, and we believe that this work is a step towards automated conjecture discovery. The project repository is available online.
