$Cover Image$

Constructive Circuit Amplification: Improving Math Reasoning in LLMs via Targeted Sub-Network Updates

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Prakash, Nikhil, Ren, Donghao, Moritz, Dominik, Assogba, Yannick
Format:	Preprint
Published:	2025
Subjects:	Computation and Language
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Similar Items

Evaluating Long Range Dependency Handling in Code Generation LLMs
by: Assogba, Yannick, et al.
Published: (2024)

Exploring Empty Spaces: Human-in-the-Loop Data Augmentation
by: Yeh, Catherine, et al.
Published: (2024)

Semantic Regexes: Auto-Interpreting LLM Features with a Structured Language
by: Boggust, Angie, et al.
Published: (2025)

EncQA: Benchmarking Vision-Language Models on Visual Encodings for Charts
by: Mukherjee, Kushin, et al.
Published: (2025)

Compress and Compare: Interactively Evaluating Efficiency and Behavior Across ML Model Compression Experiments
by: Boggust, Angie, et al.
Published: (2024)

Premise-Augmented Reasoning Chains Improve Error Identification in Math reasoning with LLMs
by: Mukherjee, Sagnik, et al.
Published: (2025)

Rectify Evaluation Preference: Improving LLMs' Critique on Math Reasoning via Perplexity-aware Reinforcement Learning
by: Tian, Changyuan, et al.
Published: (2025)

Improving Multilingual Math Reasoning for African Languages
by: Ogundepo, Odunayo, et al.
Published: (2025)

DocMath-Eval: Evaluating Math Reasoning Capabilities of LLMs in Understanding Long and Specialized Documents
by: Zhao, Yilun, et al.
Published: (2023)

rStar-Math: Small LLMs Can Master Math Reasoning with Self-Evolved Deep Thinking
by: Guan, Xinyu, et al.
Published: (2025)

Understanding and Mitigating Spurious Signal Amplification in Test-Time Reinforcement Learning for Math Reasoning
by: Yu, Yongcan, et al.
Published: (2026)

First-Step Advantage: Importance of Starting Right in Multi-Step Math Reasoning
by: Jain, Kushal, et al.
Published: (2023)

Task Complexity Matters: An Empirical Study of Reasoning in LLMs for Sentiment Analysis
by: Huang, Donghao, et al.
Published: (2026)

Right Is Not Enough: The Pitfalls of Outcome Supervision in Training LLMs for Math Reasoning
by: Guo, Jiaxing, et al.
Published: (2025)

Beyond Math: Stories as a Testbed for Memorization-Constrained Reasoning in LLMs
by: Jiang, Yuxuan, et al.
Published: (2024)

Scheherazade: Evaluating Chain-of-Thought Math Reasoning in LLMs with Chain-of-Problems
by: Miner, Stephen, et al.
Published: (2024)

MathHay: An Automated Benchmark for Long-Context Mathematical Reasoning in LLMs
by: Wang, Lei, et al.
Published: (2024)

SuperCLUE-Math6: Graded Multi-Step Math Reasoning Benchmark for LLMs in Chinese
by: Xu, Liang, et al.
Published: (2024)

Can LLMs Reason with Rules? Logic Scaffolding for Stress-Testing and Improving LLMs
by: Wang, Siyuan, et al.
Published: (2024)

CoinMath: Harnessing the Power of Coding Instruction for Math LLMs
by: Wei, Chengwei, et al.
Published: (2024)

MathGenie: Generating Synthetic Data with Question Back-translation for Enhancing Mathematical Reasoning of LLMs
by: Lu, Zimu, et al.
Published: (2024)

Can LLMs Master Math? Investigating Large Language Models on Math Stack Exchange
by: Satpute, Ankit, et al.
Published: (2024)

FinanceMath: Knowledge-Intensive Math Reasoning in Finance Domains
by: Zhao, Yilun, et al.
Published: (2023)

Formulation Comparison for Timeline Construction using LLMs
by: Hasegawa, Kimihiro, et al.
Published: (2024)

ClozeMath: Improving Mathematical Reasoning in Language Models by Learning to Fill Equations
by: Pham, Quang Hieu, et al.
Published: (2025)

The CompMath-MCQ Dataset: Are LLMs Ready for Higher-Level Math?
by: Raimondi, Bianca, et al.
Published: (2026)

Understanding Annotator Safety Policy with Interpretability
by: Oesterling, Alex, et al.
Published: (2026)

Can LLMs $\textit{understand}$ Math? -- Exploring the Pitfalls in Mathematical Reasoning
by: Roy, Tiasa Singha, et al.
Published: (2025)

Sparse Autoencoders are Capable LLM Jailbreak Mitigators
by: Assogba, Yannick, et al.
Published: (2026)

User Profile with Large Language Models: Construction, Updating, and Benchmarking
by: Prottasha, Nusrat Jahan, et al.
Published: (2025)

MathArena: Evaluating LLMs on Uncontaminated Math Competitions
by: Balunović, Mislav, et al.
Published: (2025)

AgenticMath: Enhancing LLM Reasoning via Agentic-based Math Data Generation
by: Liu, Xianyang, et al.
Published: (2025)

Does Math Reasoning Improve General LLM Capabilities? Understanding Transferability of LLM Reasoning
by: Huan, Maggie, et al.
Published: (2025)

An Empirical Study of Data Ability Boundary in LLMs' Math Reasoning
by: Chen, Zui, et al.
Published: (2024)

Using RL to Identify Divisive Perspectives Improves LLMs Abilities to Identify Communities on Social Media
by: Mehta, Nikhil, et al.
Published: (2024)

Do LLMs Overthink Basic Math Reasoning? Benchmarking the Accuracy-Efficiency Tradeoff in Language Models
by: Srivastava, Gaurav, et al.
Published: (2025)

Mitigating Catastrophic Forgetting in Target Language Adaptation of LLMs via Source-Shielded Updates
by: Yamaguchi, Atsuki, et al.
Published: (2025)

Improving Latent Reasoning in LLMs via Soft Concept Mixing
by: Wang, Kang, et al.
Published: (2025)

Reforming the Mechanism: Editing Reasoning Patterns in LLMs with Circuit Reshaping
by: Lei, Zhenyu, et al.
Published: (2026)

InternLM-Math: Open Math Large Language Models Toward Verifiable Reasoning
by: Ying, Huaiyuan, et al.
Published: (2024)