EMNLP Findings2024
Exploring Multilingual Concepts of Human Values in Large Language Models: Is Value Alignment Consistent, Transferable and Controllable across Languages?
Shaoyang Xu, Weilong Dong, Zishan Guo, Xinwei Wu, Deyi Xiong
Abstract
Prior research has revealed that certain abstract concepts are linearly represented as directions in the representation space of LLMs, predominantly centered around English. In this paper, we extend this investigation to a multilingual context, with a specific focus on human valuesrelated concepts (i.e., value concepts) due to their significance for AI safety. Through our comprehensive exploration covering 7 types of human values, 16 languages and 3 LLM series with distinct multilinguality (e.g., monolingual, bilingual and multilingual), we first empirically confirm the presence of value concepts within LLMs in a multilingual format. Further analysis on the cross-lingual characteristics of these concepts reveals 3 traits arising from language resource disparities: cross-lingual inconsistency, distorted linguistic relationships, and unidirectional cross-lingual transfer between high-and low-resource languages, all in terms of value concepts. Moreover, we validate the feasibility of cross-lingual control over value alignment capabilities of LLMs, leveraging the dominant language as a source language. Ultimately, recognizing the significant impact of LLMs' multilinguality on our results, we consolidate our findings and provide prudent suggestions on the composition of multilingual data for LLMs pre-training.
Ask Lune
Know this paper in minutes, not an afternoon.
Lune can read every page of it and the top-tier papers that cite it, and gives you its claims, its evidence and its limits, each one cited to the paper it comes from. No skimming, no AI slop.
3 top-tier papers have cited it since. Know what they found before your reviewers do.
Lune can draw on
the full paper · the top-tier papers that cite it · related work
Free to start. No credit card required.
Lineage
- DVMap: Fine-Grained Pluralistic Value Alignment via High-Consensus Demographic-Value Mapping cites Aligning AI With Shared Human Values
- Inference-Time Intervention: Eliciting Truthful Answers from a Language Model cites TruthfulQA: Measuring How Models Mimic Human Falsehoods
- Internal Value Alignment in Large Language Models through Controlled Value Vector Activation cites Inference-Time Intervention: Eliciting Truthful Answers from a Language Model
- Beyond Value Benchmarks: Measuring Value-Structure Alignment in Large Language Models via Symmetric Q-Sorts cites Aligning AI With Shared Human Values
Cited by top-tier papers3
- DVMap: Fine-Grained Pluralistic Value Alignment via High-Consensus Demographic-Value MappingPengyun Zhu, Yuqi Ren, Zhen Wang, Lei Yang et al.ACL 2026
- Internal Value Alignment in Large Language Models through Controlled Value Vector ActivationHaoran Jin, Meng Li, Xiting Wang, Zhihao Xu et al.ACL 2025
- Beyond Value Benchmarks: Measuring Value-Structure Alignment in Large Language Models via Symmetric Q-SortsJingting Zheng, Yuqi Ren, Linhao Yu, Yongqi Leng et al.ACL 2026
Builds on19
- TruthfulQA: Measuring How Models Mimic Human FalsehoodsStephanie Lin, Jacob Hilton, Owain EvansACL 2022 · 3,228 citations
- Inference-Time Intervention: Eliciting Truthful Answers from a Language ModelKenneth Li, Oam Patel, Fernanda B. Viégas, Hanspeter Pfister et al.NeurIPS 2023 · 1,549 citations
- Aligning AI With Shared Human ValuesDan Hendrycks, Collin Burns, Steven Basart, Andrew Critch et al.ICLR 2021 · 878 citations
- RealToxicityPrompts: Evaluating Neural Toxic Degeneration in Language ModelsSamuel Gehman, Suchin Gururangan, Maarten Sap, Yejin Choi et al.EMNLP Findings 2020 · 692 citations
- Multilingual Jailbreak Challenges in Large Language ModelsYue Deng, Wenxuan Zhang, Sinno Jialin Pan, Lidong BingICLR 2024 · 230 citations
Related papers
- MPO: Multilingual Safety Alignment via Reward Gap OptimizationWeixiang Zhao, Yulin Hu, Yang Deng, Tongtong Wu et al.ACL 2025 · 12 citations
- Align Once, Benefit Multilingually: Enforcing Multilingual Consistency for LLM Safety AlignmentYuyan Bu, Xiaohao Liu, ZhaoXing Ren, Yaodong Yang et al.ICLR 2026 · 9 citations
- The Language Barrier: Dissecting Safety Challenges of LLMs in Multilingual ContextsLingfeng Shen, Weiting Tan, Sihao Chen, Yunmo Chen et al.ACL Findings 2024 · 25 citations
- Multilingual Refusal Alignment for Safer Large Language ModelsAleksandra Krasnodebska, Wojciech Kusa, Aldo LipaniACL Findings 2026
- Who Transfers Safety? Identifying and Targeting Cross-Lingual Shared Safety NeuronsXianhui Zhang, Chengyu Xie, Linxia Zhu, Yonghui Yang et al.ICML 2026 · 5 citations
