MATH

The 'MATH' dataset is a benchmark containing 747,000 math problems used to evaluate the mathematical reasoning capabilities of language models.

Papers using MATH (29)

rStar-Math: Small LLMs Can Master Math Reasoning with Self-Evolved Deep Thinking2025 · 8 citesImproved Large Language Diffusion Models2026Shape of Thought: When Distribution Matters More than Correctness in Reasoning Tasks2025EvalTree: Profiling Language Model Weaknesses via Hierarchical Capability Trees2025Rewriting Pre-Training Data Boosts LLM Performance in Math and Code2025LLM Performance for Code Generation on Noisy Tasks2025Reasoning-as-Logic-Units: Scaling Test-Time Reasoning in Large Language Models Through Logic Unit Alignment2025Code-Vision: Evaluating Multimodal LLMs Logic Understanding and Code Generation Capabilities2025MathClean: A Benchmark for Synthetic Mathematical Data Cleaning2025Progressive-Hint Prompting Improves Reasoning in Large Language Models2023 · 33 citesMetaMath: Bootstrap Your Own Mathematical Questions for Large Language Models2023 · 28 citesSolving Challenging Math Word Problems Using GPT-4 Code Interpreter with Code-based Self-Verification2023 · 18 citesData Interpreter: An LLM Agent For Data Science2024 · 11 citesLearning From Mistakes Makes LLM Better Reasoner2023 · 5 citesMathCoder: Seamless Code Integration in LLMs for Enhanced Mathematical Reasoning2023 · 4 citesDotaMath: Decomposition of Thought with Code Assistance and Self-correction for Mathematical Reasoning2024 · 4 citesCREATOR: Tool Creation for Disentangling Abstract and Concrete Reasoning of Large Language Models2023 · 3 citesOpenMathInstruct-1: A 1.8 Million Math Instruction Tuning Dataset2024 · 3 citesInternLM-Math: Open Math Large Language Models Toward Verifiable Reasoning2024 · 3 citesMathGenie: Generating Synthetic Data with Question Back-translation for Enhancing Mathematical Reasoning of LLMs2024 · 2 citesMuMath-Code: Combining Tool-Use Large Language Models with Multi-perspective Data Augmentation for Mathematical Reasoning2024 · 2 citesReGAL: Refactoring Programs to Discover Generalizable Abstractions2024 · 1 citesDivide-and-Conquer Meets Consensus: Unleashing the Power of Functions in Code Generation2024 · 1 citesBuilding Math Agents with Multi-Turn Iterative Preference Learning2024 · 1 citesMARIO: MAth Reasoning with code Interpreter Output -- A Reproducible Pipeline2024Embedding Self-Correction as an Inherent Ability in Large Language Models for Enhanced Mathematical Reasoning2024ReasonAgain: Using Extractable Symbolic Programs to Evaluate Mathematical Reasoning2024UTMath: Math Evaluation with Unit Test via Reasoning-to-Coding Thoughts2024InfinityMATH: A Scalable Instruction Tuning Dataset in Programmatic Mathematical Reasoning2024