Copertina

MSCR: Exploring the Vulnerability of LLMs' Mathematical Reasoning Abilities Using Multi-Source Candidate Replacement

Fuente: arXiv

Salvato in:

Dettagli Bibliografici
Autori principali:	Sun, Zhishen, Dai, Guang, Ye, Haishan
Natura:	Preprint
Pubblicazione:	2025
Soggetti:	Artificial Intelligence
Accesso online:	Acceder al recurso
Tags:	Aggiungi Tag Nessun Tag, puoi essere il primo ad aggiungerne!!

Documenti analoghi

Numerical Sensitivity and Robustness: Exploring the Flaws of Mathematical Reasoning in Large Language Models
di: Sun, Zhishen, et al.
Pubblicazione: (2025)

ESSAM: A Novel Competitive Evolution Strategies Approach to Reinforcement Learning for Memory Efficient LLMs Fine-Tuning
di: Sun, Zhishen, et al.
Pubblicazione: (2026)

Evaluating the Reasoning Abilities of LLMs on Underrepresented Mathematics Competition Problems
di: Golladay, Samuel, et al.
Pubblicazione: (2025)

Exposing the Achilles' Heel: Evaluating LLMs Ability to Handle Mistakes in Mathematical Reasoning
di: Singh, Joykirat, et al.
Pubblicazione: (2024)

Inductive or Deductive? Rethinking the Fundamental Reasoning Abilities of LLMs
di: Cheng, Kewei, et al.
Pubblicazione: (2024)

CogMath: Assessing LLMs' Authentic Mathematical Ability from a Human Cognitive Perspective
di: Liu, Jiayu, et al.
Pubblicazione: (2025)

MAR:Multi-Agent Reflexion Improves Reasoning Abilities in LLMs
di: Ozer, Onat, et al.
Pubblicazione: (2025)

Multilingual Mathematical Reasoning: Advancing Open-Source LLMs in Hindi and English
di: Anand, Avinash, et al.
Pubblicazione: (2024)

From $O(mn)$ to $O(r^2)$: Two-Sided Low-Rank Communication for Adam in Distributed Training with Memory Efficiency
di: Dang, Sizhe, et al.
Pubblicazione: (2026)

Improving LLMs' Generalized Reasoning Abilities by Graph Problems
di: Zhang, Qifan, et al.
Pubblicazione: (2025)

Quantization Meets Reasoning: Exploring and Mitigating Degradation of Low-Bit LLMs in Mathematical Reasoning
di: Li, Zhen, et al.
Pubblicazione: (2025)

Diversity of Thought Improves Reasoning Abilities of LLMs
di: Naik, Ranjita, et al.
Pubblicazione: (2023)

RADAR: Reasoning-Ability and Difficulty-Aware Routing for Reasoning LLMs
di: Fernandez, Nigel, et al.
Pubblicazione: (2025)

ORIGAMISPACE: Benchmarking Multimodal LLMs in Multi-Step Spatial Reasoning with Mathematical Constraints
di: Xu, Rui, et al.
Pubblicazione: (2025)

Enhancing Mathematical Reasoning in LLMs with Background Operators
di: Chen, Jiajun, et al.
Pubblicazione: (2024)

FZOO: Fast Zeroth-Order Optimizer for Fine-Tuning Large Language Models towards Adam-Scale Speed
di: Dang, Sizhe, et al.
Pubblicazione: (2025)

Enhancing Reasoning Abilities of Small LLMs with Cognitive Alignment
di: Cai, Wenrui, et al.
Pubblicazione: (2025)

Navigating the Labyrinth: Evaluating LLMs' Ability to Reason About Search Problems
di: Borazjanizadeh, Nasim, et al.
Pubblicazione: (2024)

Can LLMs Reliably Simulate Real Students' Abilities in Mathematics and Reading Comprehension?
di: Srivatsa, KV Aditya, et al.
Pubblicazione: (2025)

Reasoning with LLMs for Zero-Shot Vulnerability Detection
di: Zibaeirad, Arastoo, et al.
Pubblicazione: (2025)

How Trustworthy are Open-Source LLMs? An Assessment under Malicious Demonstrations Shows their Vulnerabilities
di: Mo, Lingbo, et al.
Pubblicazione: (2023)

Can MLLMs Absorb Math Reasoning Abilities from LLMs as Free Lunch?
di: Hu, Yijie, et al.
Pubblicazione: (2025)

Improving Multimodal LLMs Ability In Geometry Problem Solving, Reasoning, And Multistep Scoring
di: Anand, Avinash, et al.
Pubblicazione: (2024)

Boosting Vulnerability Detection of LLMs via Curriculum Preference Optimization with Synthetic Reasoning Data
di: Wen, Xin-Cheng, et al.
Pubblicazione: (2025)

Embedding Self-Correction as an Inherent Ability in Large Language Models for Enhanced Mathematical Reasoning
di: Gao, Kuofeng, et al.
Pubblicazione: (2024)

SAAS: Solving Ability Amplification Strategy for Enhanced Mathematical Reasoning in Large Language Models
di: Kim, Hyeonwoo, et al.
Pubblicazione: (2024)

HERMES: Towards Efficient and Verifiable Mathematical Reasoning in LLMs
di: Ospanov, Azim, et al.
Pubblicazione: (2025)

Weight Patching: Toward Source-Level Mechanistic Localization in LLMs
di: Sun, Chenghao, et al.
Pubblicazione: (2026)

Guarded Repair for Harm-Aware Post-hoc Replacement of LLM Mathematical Reasoning
di: Xia, Haizhou
Pubblicazione: (2026)

QED: An Open-Source Multi-Agent System for Generating Mathematical Proofs on Open Problems
di: An, Chenyang, et al.
Pubblicazione: (2026)

Harnessing the Power of LLMs in Source Code Vulnerability Detection
di: Mahyari, Andrew A
Pubblicazione: (2024)

Improving Multi-Step Reasoning Abilities of Large Language Models with Direct Advantage Policy Optimization
di: Liu, Jiacai, et al.
Pubblicazione: (2024)

Beyond Accuracy: Dissecting Mathematical Reasoning for LLMs Under Reinforcement Learning
di: Wang, Jiayu, et al.
Pubblicazione: (2025)

ReasonAgain: Using Extractable Symbolic Programs to Evaluate Mathematical Reasoning
di: Yu, Xiaodong, et al.
Pubblicazione: (2024)

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

Evaluating the Logical Reasoning Abilities of Large Reasoning Models
di: Liu, Hanmeng, et al.
Pubblicazione: (2025)

Exploring ChatGPT's Empathic Abilities
di: Schaaff, Kristina, et al.
Pubblicazione: (2023)

A Survey of Multimodal Mathematical Reasoning: From Perception, Alignment to Reasoning
di: Yang, Tianyu, et al.
Pubblicazione: (2026)

Unmasking Reasoning Processes: A Process-aware Benchmark for Evaluating Structural Mathematical Reasoning in LLMs
di: Zheng, Xiang, et al.
Pubblicazione: (2026)

Plan before Solving: Problem-Aware Strategy Routing for Mathematical Reasoning with LLMs
di: Qi, Shihao, et al.
Pubblicazione: (2025)