Challenge: Large language models (LLMs) are often evaluated on math word problems . however, such metrics conflate two distinct sub-skills: abstract formulation and arithmetic computation.
Approach: They propose to use Final-answer-based metrics to evaluate large language models on math word problems to conflate two distinct sub-skills: abstract formulation and arithmetic computation.
Outcome: The proposed model performance is bottlenecked by arithmetic computation and not abstract formulation, the study shows.

Similar Papers

Over-Reasoning and Redundant Calculation of Large Language Models (2024.eacl-short)

Copied to clipboard

Challenge: Large language models (LLMs) can solve problems step-by-step, but it is unclear whether they know when to use CoT and whether they are always necessary.
Approach: They propose to use LLMs to generate redundant calculations and reasoning on a manually constructed math QA dataset, GSM8K-Zero.
Outcome: The proposed model generates redundant calculations and reasoning on a manually constructed math QA dataset, but it is unclear whether it is necessary to use CoT reasoning.
LLMs Faithfully and Iteratively Compute Answers During CoT: A Systematic Analysis With Multi-step Arithmetics (2026.findings-eacl)

Copied to clipboard

Challenge: Specifically, we examine when the LLMs’ answer is (pre)determined, especially before the CoT begins or after, and how strongly the information from CoT specifically has a causal effect on the final answer.
Approach: They examine when the LLMs’ answer is (pre)determined, especially before the CoT begins or after, and how strongly the information from CoT specifically has a causal effect on the final answer.
Outcome: The proposed model can generate reasoning chains while generating the reasoning chain on the fly.
Memory or Reasoning? Explore How LLMs Compute Mixed Arithmetic Expressions (2025.findings-acl)

Copied to clipboard

Challenge: Large language models (LLMs) can solve complex multi-step math reasoning problems, but their internal implementation is limited.
Approach: They propose to use a "C**ausal **E**ffect **D**riven **F**ine-tuning method" to improve LLMs' reasoning ability.
Outcome: The proposed method improves the model's reasoning ability by enhancing key components that are used to execute mixed arithmetic calculations.
Disentangling Text and Math in Word Problems: Evidence for the Bidimensional Structure of Large Language Models’ Reasoning (2025.findings-acl)

Copied to clipboard

Challenge: Existing studies show that LLMs struggle with text interpretation and equation solving, despite distinct proficiencies in textual and mathematical components.
Approach: They disentangle textual interpretation and mathematical solving steps in word problems drawn from Brazil's largest college entrance exam and popular grade school-level benchmark GSM8K.
Outcome: The proposed model outperforms LLMs in Brazil's largest college entrance exam and popular grade school-level benchmark.
Arithmetic Reasoning with LLM: Prolog Generation & Permutation (2024.naacl-short)

Copied to clipboard

Challenge: Existing work has shown that large language models can generate arithmetic and commonsense reasoning, but they are not native to mathematical operations and symbolic manipulations.
Approach: They propose to use large language models to generate Prolog programs to solve math problems using a code interpreter to generate arithmetic and symbolic formulas.
Outcome: The proposed model outperforms CoT generation in the GSM8K benchmark across three LLMs.
Do LLMs Overthink Basic Math Reasoning? Benchmarking the Accuracy-Efficiency Tradeoff in Language Models (2026.findings-acl)

Copied to clipboard

Challenge: Large language models (LLMs) achieve impressive performance on complex benchmarks yet sometimes fail on basic math reasoning.
Approach: They propose a benchmark to evaluate the efficiency of reasoning in large language models . they formalize the accuracy-verbosity tradeoff and introduce the overthinking score .
Outcome: The proposed model performs well on complex benchmarks but fails on basic math reasoning . the proposed model generates 18 more tokens while achieving lower accuracy .
LLMs for Mathematical Modeling: Towards Bridging the Gap between Natural and Mathematical Languages (2025.findings-naacl)

Copied to clipboard

Challenge: Large Language Models (LLMs) have demonstrated strong performance across various natural language processing tasks, but their proficiency in mathematical reasoning remains a key challenge.
Approach: They propose a process-oriented framework to evaluate LLMs' ability to construct mathematical models, using solvers to compare outputs with ground truth.
Outcome: The proposed framework evaluates LLMs' ability to construct mathematical models, using solvers to compare outputs with ground truth.
Exposing the Achilles’ Heel: Evaluating LLMs Ability to Handle Mistakes in Mathematical Reasoning (2025.acl-long)

Copied to clipboard

Challenge: Existing evaluations focus on final accuracy, neglecting the critical aspect of reasoning capabilities.
Approach: They propose to evaluate LLMs’ abilities to detect and correct reasoning mistakes by using rule-based methods and smaller language models.
Outcome: The proposed model outperforms existing models such as GPT-4o and GPT4 in both accuracy and accuracy, but lacks data contamination and memorization concerns.
Interpretability Analysis of Arithmetic In-Context Learning in Large Language Models (2025.emnlp-main)

Copied to clipboard

Challenge: Large language models (LLMs) solve arithmetic with only a few in-context examples, yet the computations that connect those examples to the answer remain opaque.
Approach: They propose to use in-context examples to illustrate how large language models process ICEs to isolate partial-sum representations in three-operand tasks and investigate their influence on final logits.
Outcome: The proposed model performs better than previous models on three-operand tasks.
PyraMathBench: Evaluating and Improving Mathematical Capability in Large Language Models (2026.findings-acl)

Copied to clipboard

Challenge: Numerical reasoning is ubiquitous in scientific research and financial analysis, but few benchmarks evaluate them by integrating numerical processing and mathematical reasoning.
Approach: They propose a numerically-integrated hierarchical benchmark with 27,215 questions derived from 7,404 math word problems that spans 4 key cognitive aspects, 14 subcategories, and 2 modalities.
Outcome: The proposed model improves Qwen-2.5 score with SOLVE and IRPO training.

What is GenGO?

GenGO is an NLP powered publication search system. It currenctly indexes 30k+ papers from ACL Anthology, and implements multi-aspect summarization, semantic search, and more!

Information

About
Limitations