AI Paper Atlas

Research map / Large language models / Agents, code and math

AgentMath: Empowering Mathematical Reasoning for Large Language Models via Tool-Augmented Agent

Haipeng Luo, Huawen Feng, Qingfeng Sun, Can Xu, Kai Zheng, Yufei Wang, TAO YANG, Han Hu, Yansong Tang

ICLR 2026 · Large language models › Agents, code and math · approach: Inductive and abstract reasoning evaluation · 15 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
use tool-augmented synthetic data instead of text-only data in SFT (20k) (ablation)Qwen3-8B-Base SFT · AIME24accuracy: 57.1% -> 60.5%Improved
use tool-augmented synthetic data instead of text-only data in SFT (20k) (ablation)Qwen3-8B-Base SFT · AIME25accuracy: 49.2% -> 53.3%Improved
add agentic RL with code tools instead of text-based RL (ablation)Qwen3-8B-Base · AIME24accuracy: 68.7% -> 76.2%Improved
add agentic RL with code tools instead of text-based RL (ablation)Qwen3-8B-Base · AIME25accuracy: 57.5% -> 67.5%Improved
add tool augmentation in RL for training efficiency (ablation)Qwen3-8B-Base RL · RL steps to convergesteps: ~1600 -> ~400 stepsImproved
add format consistency correction to synthetic data (ablation)AgentMath SFT data (20k) · AIME24accuracy: 35.3% -> 47.4%Improved
add code executability verification (ablation)AgentMath SFT data (20k) · AIME24accuracy: 47.4% -> 52.8%Improved
add environmental feedback alignment (ablation)AgentMath SFT data (20k) · AIME24accuracy: 52.8% -> 56.3%Improved
add self-correction capability injection (ablation)AgentMath SFT data (20k) · AIME24accuracy: 56.3% -> 58.6%Improved
add SFT with selective feedback masking of interpreter outputs (ablation)AgentMath SFT data (20k) · AIME24accuracy: 58.6% -> 60.5%Improved
add SFT with selective feedback masking of interpreter outputs (ablation)AgentMath SFT data (20k) · AIME25accuracy: 50.8% -> 53.3%Improved
scale tool-augmented SFT data from 2k to 300k (ablation)AgentMath SFT · AIME24accuracy: 27.2% -> 78.4%Improved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy