AI Skill Hub 强烈推荐:数学研究代理 是一款优质的Agent工作流。AI 综合评分 8.0 分,在同类工具中表现稳健。如果你正在寻找可靠的Agent工作流解决方案,这是一个值得深入了解的选择。
数学研究代理 是一套完整的 AI Agent 自动化工作流方案。通过可视化的节点编排,将复杂的多步骤任务拆解为清晰的自动化流程,实现全程无人值守的智能处理。支持与数百种外部服务和 API 无缝集成,适合构建数据处理管线、业务自动化和 AI 辅助决策系统。
数学研究代理 是一套完整的 AI Agent 自动化工作流方案。通过可视化的节点编排,将复杂的多步骤任务拆解为清晰的自动化流程,实现全程无人值守的智能处理。支持与数百种外部服务和 API 无缝集成,适合构建数据处理管线、业务自动化和 AI 辅助决策系统。
# 克隆仓库 git clone https://github.com/sjtuytc/ResearchMathAgent cd ResearchMathAgent # 查看安装说明 cat README.md # 按 README 完成环境依赖安装后即可使用
# 查看帮助 researchmathagent --help # 基本运行 researchmathagent [options] <input> # 详细使用说明请查阅文档 # https://github.com/sjtuytc/ResearchMathAgent
# researchmathagent 配置说明 # 查看配置选项 researchmathagent --config-example > config.yml # 常见配置项 # output_dir: ./output # log_level: info # workers: 4 # 环境变量(覆盖配置文件) export RESEARCHMATHAGENT_CONFIG="/path/to/config.yml"
Official code release for RMA. RMA is a research math agent system that turns problem statements into verifiable proof artifacts.
initializer -> proposer -> verifier -> refiner).We present Research Math Agents (RMA), an agentic framework for automated reasoning on research-level mathematical problems. Unlike prior studies centered on competition mathematics or formal theorem proving, RMA targets research-level mathematical problems that require long-horizon reasoning, literature grounding, and iterative proof refinement. RMA decomposes research-level proof solving into specialized modules for problem analysis, literature search and understanding, fair comparison, knowledge-bank construction, and proof verification, all coordinated by initializer, proposer, and verifier agents through a shared structured memory. Within this unified framework, these agents operate in a multi-role, multi-round workflow, collaboratively generating, refining, and verifying candidate proofs through iterative feedback. We evaluate RMA on the First Proof benchmark, which consists of ten research-level problems contributed by expert mathematicians across diverse domains. Through comprehensive expert evaluation, RMA outperforms strong baselines on the First Proof benchmark, including GPT-5.2R and Aletheia, solving eight out of ten research problems and producing more logically sound and readable proofs. Our comprehensive ablation studies further show that performance gains arise from the interaction of structured reasoning modules, iterative refinement, and verifier-based feedback, rather than any single component.

---

RMA targets research-level mathematics (not just competition math or formal theorem proving) by combining specialized modules for: - problem analysis, - literature search and understanding, - fair comparison, - knowledge-bank construction, and - proof verification.
Within a multi-role, multi-round workflow, initializer/proposer/verifier agents share structured memory to iteratively generate, refine, and validate candidate proofs. On the First Proof benchmark, RMA reports stronger results than strong baselines through structured modules, iterative refinement, and verifier feedback.
---
To compile the paper:
latexmk -pdf -interaction=nonstopmode -halt-on-error main.tex
---
The executable pipeline is:
parse -> propose -> verify -> refine
Each stage can be run directly. Later stages automatically initialize missing earlier artifacts in the same run folder.
rma parse q6
rma propose q6
rma verify q6
rma refine q6
All four stage commands accept the same experiment/model folder controls:
rma parse q6 --exp-name proofs_v1_june13 --model-name rma-skeleton
rma propose q6 --exp-name proofs_v1_june13 --model-name rma-skeleton
rma verify q6 --exp-name proofs_v1_june13 --model-name rma-skeleton
rma refine q6 --exp-name proofs_v1_june13 --model-name rma-skeleton
Stage outputs:
rma parse: copies the problem file and writes parsed_problem.json plus problem_analysis.md.rma propose: writes qN_solution.tex and versioned proposal artifacts.rma verify: checks the current solution, renders PDF by default, and writes verification reports. Verification includes LaTeX/artifact checks and mathematical-completeness gates for proof length, subclaim structure, subproofs, theorem hypothesis audits, citations or derivations, and boundary-case proofs.rma refine: consumes the latest verification report and rewrites the current solution only when issues were found.高质量的AI工作流,适用于数学研究
该工具未明确声明开源协议,商业使用前请联系原作者确认授权范围,避免侵权风险。
AI Skill Hub 为第三方内容聚合平台,本页面信息基于公开数据整理,不对工具功能和质量作任何法律背书。
建议在沙箱或测试环境中充分验证后,再部署至生产环境,并做好必要的安全评估。
总体来看,数学研究代理 是一款质量优秀的Agent工作流,在同类工具中具备一定竞争力。AI Skill Hub 将持续追踪其更新动态,建议收藏备用,结合自身场景选择合适时机引入使用。
| 原始名称 | ResearchMathAgent |
| 原始描述 | 开源AI工作流:Research Math Agents, Official code release for our paper RMA。⭐17 · TeX |
| Topics | aiai-agentsmathscience |
| GitHub | https://github.com/sjtuytc/ResearchMathAgent |
| 语言 | TeX |
收录时间:2026-06-16 · 更新时间:2026-06-16 · License:未公布 · AI Skill Hub 不对第三方内容的准确性作法律背书。
选择 Agent 类型,复制安装指令后粘贴到对应客户端