Cover Image

How Does Alignment Enhance LLMs' Multilingual Capabilities? A Language Neurons Perspective

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Zhang, Shimao, Lai, Zhejian, Liu, Xiang, She, Shuaijie, Liu, Xiao, Gong, Yeyun, Huang, Shujian, Chen, Jiajun
Format:	Preprint
Published:	2025
Subjects:	Computation and Language Artificial Intelligence
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Similar Items

MAPO: Advancing Multilingual Reasoning through Multilingual Alignment-as-Preference Optimization
by: She, Shuaijie, et al.
Published: (2024)

Multilingual Contrastive Decoding via Language-Agnostic Layers Skipping
by: Zhu, Wenhao, et al.
Published: (2024)

TAPO: Translation Augmented Policy Optimization for Multilingual Mathematical Reasoning
by: Huang, Xu, et al.
Published: (2026)

Question Translation Training for Better Multilingual Reasoning
by: Zhu, Wenhao, et al.
Published: (2024)

Process-based Self-Rewarding Language Models
by: Zhang, Shimao, et al.
Published: (2025)

Exploring the Factual Consistency in Dialogue Comprehension of Large Language Models
by: She, Shuaijie, et al.
Published: (2023)

Alleviating Distribution Shift in Synthetic Data for Machine Translation Quality Estimation
by: Geng, Xiang, et al.
Published: (2025)

R-PRM: Reasoning-Driven Process Reward Modeling
by: She, Shuaijie, et al.
Published: (2025)

A Data-Efficient Path to Multilingual LLMs: Language Expansion via Post-training PARAM$Δ$ Integration into Upcycled MoE
by: Zhou, Hao, et al.
Published: (2026)

PATS: Process-Level Adaptive Thinking Mode Switching
by: Wang, Yi, et al.
Published: (2025)

EDT: Improving Large Language Models' Generation by Entropy-based Dynamic Temperature Sampling
by: Zhang, Shimao, et al.
Published: (2024)

Getting More from Less: Large Language Models are Good Spontaneous Multilingual Learners
by: Zhang, Shimao, et al.
Published: (2024)

Understanding New-Knowledge-Induced Factual Hallucinations in LLMs: Analysis and Interpretation
by: Dang, Renfei, et al.
Published: (2025)

PreAlign: Boosting Cross-Lingual Transfer by Early Establishment of Multilingual Alignment
by: Li, Jiahuan, et al.
Published: (2024)

Self-Improving Multilingual Long Reasoning via Translation-Reasoning Integrated Training
by: Liu, Junxiao, et al.
Published: (2026)

Why Not Transform Chat Large Language Models to Non-English?
by: Geng, Xiang, et al.
Published: (2024)

Multilingual Pretraining and Instruction Tuning Improve Cross-Lingual Knowledge Alignment, But Only Shallowly
by: Gao, Changjiang, et al.
Published: (2024)

Eliciting the Translation Ability of Large Language Models via Multilingual Finetuning with Translation Instructions
by: Li, Jiahuan, et al.
Published: (2023)

Lost in the Source Language: How Large Language Models Evaluate the Quality of Machine Translation
by: Huang, Xu, et al.
Published: (2024)

Multilingual Machine Translation with Large Language Models: Empirical Results and Analysis
by: Zhu, Wenhao, et al.
Published: (2023)

Language-Specific Neurons: The Key to Multilingual Capabilities in Large Language Models
by: Tang, Tianyi, et al.
Published: (2024)

Trans-Zero: Self-Play Incentivizes Large Language Models for Multilingual Translation Without Parallel Data
by: Zou, Wei, et al.
Published: (2025)

From Neurons to Semantics: Evaluating Cross-Linguistic Alignment Capabilities of Large Language Models via Neurons Alignment
by: Huang, Chongxuan, et al.
Published: (2025)

Ensembling Large Language Models with Process Reward-Guided Tree Search for Better Complex Reasoning
by: Park, Sungjin, et al.
Published: (2024)

How Does Quantization Affect Multilingual LLMs?
by: Marchisio, Kelly, et al.
Published: (2024)

SLAM: Towards Efficient Multilingual Reasoning via Selective Language Alignment
by: Fan, Yuchun, et al.
Published: (2025)

The Power of Question Translation Training in Multilingual Reasoning: Broadened Scope and Deepened Insights
by: Zhu, Wenhao, et al.
Published: (2024)

Overcoming Vocabulary Mismatch: Vocabulary-agnostic Teacher Guided Language Modeling
by: Shin, Haebin, et al.
Published: (2025)

LLMs Beyond English: Scaling the Multilingual Capability of LLMs with Cross-Lingual Feedback
by: Lai, Wen, et al.
Published: (2024)

Align to the Pivot: Dual Alignment with Self-Feedback for Multilingual Math Reasoning
by: Zhao, Chunxu, et al.
Published: (2026)

Understanding LLMs' Cross-Lingual Context Retrieval: How Good It Is And Where It Comes From
by: Gao, Changjiang, et al.
Published: (2025)

MoE-LPR: Multilingual Extension of Large Language Models through Mixture-of-Experts with Language Priors Routing
by: Zhou, Hao, et al.
Published: (2024)

Formality is Favored: Unraveling the Learning Preferences of Large Language Models on Data with Conflicting Knowledge
by: Li, Jiahuan, et al.
Published: (2024)

Enhancing Large Language Model Performance with Gradient-Based Parameter Selection
by: Li, Haoling, et al.
Published: (2024)

Multilingual != Multicultural: Evaluating Gaps Between Multilingual Capabilities and Cultural Alignment in LLMs
by: Rystrøm, Jonathan, et al.
Published: (2025)

MT-PATCHER: Selective and Extendable Knowledge Distillation from Large Language Models for Machine Translation
by: Li, Jiahuan, et al.
Published: (2024)

Extend Adversarial Policy Against Neural Machine Translation via Unknown Token
by: Zou, Wei, et al.
Published: (2025)

Friend or Foe: How LLMs' Safety Mind Gets Fooled by Intent Shift Attack
by: Ding, Peng, et al.
Published: (2025)

How Language Models Process Negation
by: Zhou, Zhejian, et al.
Published: (2026)

Improved Paraphrase Generation via Controllable Latent Diffusion
by: Zou, Wei, et al.
Published: (2024)