paper-with-me

Papers

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

2025-12-18 · Nikhil Prakash, Donghao Ren, Dominik Moritz, Yannick Assogba arxiv

Prior studies investigating the internal workings of LLMs have uncovered sparse subnetworks, often referred to as circuits, that are responsible for performing specific tasks. Additionally, it has been shown that model performance improvement through fine-tuning often results from the strengthening of existing circuits in the model. Taken together, these findings suggest the possibility of intervening directly on such circuits to make precise, task-targeted updates. Motivated by these findings, we propose a novel method called Constructive Circuit Amplification which identifies pivotal tokens from model reasoning traces as well as model components responsible for the desired task, and updates only those components. Applied to mathematical reasoning, it improves accuracy by up to +11.4% across multiple models while modifying as little as 1.59% of model components, with minimal impact on other abilities as measured by MMLU, TriviaQA, and TruthfulQA. These results demonstrate that targeted capabilities can be reliably enhanced by selectively updating a sparse set of model components.

📄 PDF Abstract BibTeX arXiv:2512.16914

Code (0)

등록된 구현이 없습니다.

Tasks

Mathematical Reasoning

Similar Papers 제목 키워드 기반

SAAS: Solving Ability Amplification Strategy for Enhanced Mathematical Reasoning in Large Language Models

2024-04-05 · Hyeonwoo Kim, Gyoungjin Gim, Yungi Kim, Jihoo Kim 외

This study presents a novel learning approach designed to enhance both mathematical reasoning and problem-solving abilities of Large Language Models (LLMs). We focus on integrating the Chain-of-Thought (CoT) and the Prog…

Mathematical Reasoning

CircuitSeer: Mining High-Quality Data by Probing Mathematical Reasoning Circuits in LLMs

2025-10-21 · Shaobo Wang, Yongliang Miao, Yuancheng Liu, Qianli Ma 외 arxiv

Large language models (LLMs) have demonstrated impressive reasoning capabilities, but scaling their performance often relies on massive reasoning datasets that are computationally expensive to train on. Existing data sel…

Mathematical Reasoning

Unmasking Reasoning Processes: A Process-aware Benchmark for Evaluating Structural Mathematical Reasoning in LLMs

2026-01-31 · Xiang Zheng, Weiqi Zhai, Wei Wang, Boyu Yang 외 arxiv

Recent large language models (LLMs) achieve near-saturation accuracy on many established mathematical reasoning benchmarks, raising concerns about their ability to diagnose genuine reasoning competence. This saturation l…

Mathematical Reasoning

DeepMath-Creative: A Benchmark for Evaluating Mathematical Creativity of Large Language Models

2025-05-13 · Xiaoyang Chen, Xinan Dai, Yu Du, Qian Feng 외

To advance the mathematical proficiency of large language models (LLMs), the DeepMath team has launched an open-source initiative aimed at developing an open mathematical LLM and systematically evaluating its mathematica…

Beyond Lines and Circles: Unveiling the Geometric Reasoning Gap in Large Language Models

2024-02-06 · Spyridon Mouselinos, Henryk Michalewski, Mateusz Malinowski

Large Language Models (LLMs) demonstrate ever-increasing abilities in mathematical and algorithmic tasks, yet their geometric reasoning skills are underexplored. We investigate LLMs' abilities in constructive geometric p…

Mathematical ReasoningVariable Selection