arXiv Machine Learning

When Safety Speaks a Language: A Mechanistic Analysis of Safety-Language Identity Entanglement in LLMs

arXiv Machine Learning
Sep 22

Multilingual Safety Signals Are Multi-Layered: Filtering Safety-Degrading Data for Safer LLMs

The paper introduces MMSAFE, a multi-layer framework designed to identify safety-degrading data in multilingual large language models. It shows that safety signals are distributed across multiple layers and only partially shared across languages, unlike the single-layer assumption used in monolingual settings. Experiments demonstrate that MMSAFE reduces harmful-response rates by 60% compared to random filtering and outperforms the best single-layer baseline across various models, languages, and safety benchmarks.

By Jiakun Li, Guowei Song, Sijia Li, Xingwei He, Hongzheng Chai, Yuan Yuan
arXiv AI
Aug 18

LLM Safety Alignment in Low-Resource Languages: A Systematic Literature Review

arXiv:2608. 14626v1 Announce Type: cross Abstract: Large Language Models (LLMs) have achieved substantial progress in safety alignment, yet their safety guarantees remain significantly weaker in low-resource and multilingual settings than in high-resource languages.

By Valdini Douglace Lemofouet, Blessing Ngozi Uzor, Paula Chikaodinaka Anyanwu, Danielle Blanche Kapsa, Sukairaj Hafiz Imam, P Sam Sahil, Abigail Oppong, Tassallah Abdullahi, Clemencia Siro, Idris Abdulmumin, Seid Muhie Yimam, Shamsuddeen Hassan Muhammad
Hugging Face Trending Papers
Jul 2

Safety Targeted Embedding Exploit via Refinement

Safety training for large language models (LLMs) is conducted predominantly in English, leaving uncertain how well safety mechanisms generalize to low-resource languages and mixed-language code-switching. We show that this creates an epistemic gap in which models confidently generate harmful responses for inputs that fall outside the distribution of their safety training.

arXiv Computation and Language
Oct 16

PRISM: Transfer and Limits of Activation Interventions in Tool-Calling Language Models

PRISM is a framework that evaluates how activation interventions transfer across precision, languages, and model scales to mitigate unsafe tool‑calling behaviors in language models. By leveraging lower‑resource settings such as quantized NF4 labels and cross‑lingual transfer, PRISM demonstrates that alignment improvements can be achieved with minimal scenarios and that these gains transfer to held‑out languages and larger models. The study shows that auxiliary supervision from compressed models and multilingual data can effectively uncover and reduce hazardous tool‑calling behaviors across different model families.

By Giovana Kerche Bon\'as, Thales Sales Almeida, Marcelo da Silva Reis, Rodrigo Nogueira, Marcos Meideiros Raimundo