Mathematical reasoning

1. This week · 2
2. 7 OctWhich Language Should a Skeleton Speak? Language Choices in Multilingual Reasoning also fits: Reasoning benchmarks and evaluation
3. 6 OctAlgorithmic Scratchpads and Curriculum Staging for Arithmetic Reasoning in Tiny Transformers
4. Earlier · 15
5. 4 OctWhen Does Longer Reasoning Help? Predicting Mathematical Reasoning Through Discovery and Execution also fits: Test-time compute and efficient thinking
6. 3 OctDense Neuro-Symbolic Reasoning in a Unified Geometry State also fits: Spatial and embodied reasoning
7. 1 OctThe Missing Primitive: Diagnosing and Repairing Mathematical Reasoning in Large Language Models
8. 30 SepReason in Style: Discovering and Controlling Style in Language Models also fits: Reasoning benchmarks and evaluation
9. 30 SepANI: Adaptive Numerical Injection for Unifying Semantic and Arithmetic Representations in Numerical Reasoning
10. 29 SepMathematical Transfer in LLMs Follows Reasoning Approach More Than Topic
11. 29 SepCorrect Answers, Invalid Traces: What Verifiable Grade-School Math Reveals About Chain-of-Thought Traces also fits: Prompting and chain-of-thought reasoning
12. 29 SepBeyond Compression: Diagnosing How Post-Training Changes Mathematical Reasoning also fits: Test-time compute and efficient thinking

1. DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models2024cited by 142 here
2. Solving Quantitative Reasoning Problems with Language Models2022cited by 116 here
3. WizardMath: Empowering Mathematical Reasoning for Large Language Models via Reinforced Evol-Instruct2023cited by 99 here
4. ToRA: A Tool-Integrated Reasoning Agent for Mathematical Problem Solving2023cited by 68 here
5. Llemma: An Open Language Model For Mathematics2023cited by 68 here

Each paper on the line cites the one before.

1. 2022Chain-of-Thought Prompting Elicits Reasoning in Large Language Modelscited by 281 here
2. 2022Self-Consistency Improves Chain of Thought Reasoning in Language Modelscited by 172 here
3. 2022STaR: Bootstrapping Reasoning With Reasoningcited by 52 here
4. 2022Large Language Models are Zero-Shot Reasonerscited by 110 here
5. 2022PAL: Program-aided Language Modelscited by 80 here
6. 2023Self-Refine: Iterative Refinement with Self-Feedbackcited by 49 here
7. 2023Tree of Thoughts: Deliberate Problem Solving with Large Language Modelscited by 87 here
8. 2023Scaling Relationship on Learning Mathematical Reasoning with Large Language Modelscited by 51 here
9. 2023WizardMath: Empowering Mathematical Reasoning for Large Language Models via Reinforced Evol-Instructcited by 99 here
10. 2023ToRA: A Tool-Integrated Reasoning Agent for Mathematical Problem Solvingcited by 68 here
11. 2023Math-Shepherd: Verify and Reinforce LLMs Step-by-step without Human Annotationscited by 60 here
12. 2024DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Modelscited by 142 here
13. 2025DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learningcited by 125 here

The 24 most cited of 1,441.

1. 6GSM-Symbolic: Understanding the Limitations of Mathematical Reasoning in Large Language Models202458
2. 7Scaling Relationship on Learning Mathematical Reasoning with Large Language Models202351
3. 8Large Language Models for Mathematical Reasoning: Progresses and Challenges202447
4. 9MathScale: Scaling Instruction Tuning for Mathematical Reasoning202442
5. 10Omni-MATH: A Universal Olympiad Level Mathematic Benchmark For Large Language Models202442
6. 11Generative Language Modeling for Automated Theorem Proving202040
7. 12MathCoder: Seamless Code Integration in LLMs for Enhanced Mathematical Reasoning202337
8. 13OpenMathInstruct-1: A 1.8 Million Math Instruction Tuning Dataset202435
9. 14DeepSeek-Prover-V2: Advancing Formal Mathematical Reasoning via Reinforcement Learning for Subgoal Decomposition202533
10. 15MathGenie: Generating Synthetic Data with Question Back-translation for Enhancing Mathematical Reasoning of LLMs202431