arXiv Computation and Language October 12, 2026 By Pardis Sadat Zahraei, Gokhan Tur, Dilek Hakkani-T\"ur, Ehsaneddin Asgari

The Alignment Veto: How Safety Training Suppresses Cultural Knowledge in LLMs

Read the original on arXiv Computation and Language →

The paper investigates how alignment training in large language models (LLMs) can suppress cultural knowledge rather than erase it. By surveying 1.53 million humans across 16 MENA countries and testing 26 models, the authors find that when models refuse to answer, their internal logit distributions still align closely with human responses, a phenomenon they term the "alignment veto." They distinguish between suppression failures—where accurate internal representations are blocked at output—and representational bias failures—where the encoding itself diverges from human values, and show that these require different interventions. The study also reveals significant inequities in the safety tax, with a 37.6 % cost and a 19.8 % alignment-quality gap between best- and worst-served nations, and notes that native-language prompting actually widens this gap. Sparse autoencoder analysis identifies a candidate feature in the DPO stage that mediates suppression in the Tulu-3-8B model.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv Computation and Language.

arXiv Computation and Language

Sep 2

Aligned but Flattened: Analyzing the Trade-off between Cultural Alignment and Diversity in LLMs

arXiv:2609.00565v1 Announce Type: cross Abstract: Cultural fine-tuning has become the de facto paradigm for building culture-aware large language models (LLMs), yet existing optimization exclusively...

By Jingshen Zhang, Shaoyang Xu, Wenxuan Zhang

llms fine-tuning benchmarks safety

Hugging Face Trending Papers

Jun 17

The Wrong Kind of Right: Quantifying and Localizing Misfired Alignment in LLMs

Warning: This paper studies stereotypes and biases, and contains potentially disturbing examples, used for illustration purposes only. Our findings should not be interpreted as an argument against alignment.

llms benchmarks safety

arXiv Computation and Language

Sep 21

Cultural Alignment in Large Language Models Using Soft Prompt Tuning

The paper proposes a method for culturally aligning large language models (LLMs) using soft prompt tuning optimized via Differential Evolution (DE). Unlike traditional fine‑tuning or reinforcement learning, this approach keeps model weights frozen and requires no preference data, instead leveraging aggregated survey scores from Hofstede's Value Survey Module (VSM13). Experiments on four countries and four instruction‑tuned models show that DE‑optimized prompts reduce cultural discrepancy, improve agreement with the World Values Survey, and are preferred in blinded pairwise evaluations by LLM judges.

By Reem I. Masoud, Martin Ferianc, Philip Treleaven, Miguel Rodrigues

llms reinforcement-learning fine-tuning safety

arXiv AI

Sep 1

Measurement Validity in LLM Cultural Alignment

arXiv:2608.29266v1 Announce Type: cross Abstract: Researchers increasingly treat LLM survey responses as a proxy for human cultural values. This includes projecting model outputs onto instruments lik...

By An Duy Nguyen, Muhammad Aurangzeb Ahmad

llms safety

arXiv AI

Sep 30

Population Fidelity: Evaluating Population Representativeness in LLMs

The paper introduces Population Fidelity, an evaluation framework for assessing how well large language models (LLMs) represent human population attitudes. It focuses on three dimensions: group-level accuracy, between-group variation, and the structure of that variation. Using the framework, the authors replicate a prior study on machine bias and test cultural fine-tuning, finding that while fine-tuning improves overall alignment, it does not enhance representation of within-population differences.

By Neemias B. da Silva, Martin Lukk, Ali Sutani, Abhishek Moturu, Harris Yang, Daniel Silver, Matt Ratto, Thiago H. Silva

llms fine-tuning safety

Hugging Face Trending Papers

Jul 9

PLURAL: A Global Dataset for Value Alignment

Large language models (LLMs) are used worldwide, yet disproportionately reflect Western values, limiting their ability to represent diverse value systems. We introduce PLURAL, a large-scale, value-focused preference dataset grounded in the Integrated Values Survey (IVS), a nationally representative survey spanning 92 countries.