Soteria: Language-Specific Functional Parameter Steering for Multilingual Safety Alignment
Ensuring consistent safety across multiple languages remains a significant challenge for large language models (LLMs). We introduce Soteria, a lightweight yet powerful strategy that locates and minimally adjusts the "functional heads" most responsible for harmful content generation in each language. By altering only a fraction of parameters, Soteria drastically reduces policy violations without sacrificing overall model performance, even in low-resource settings. To rigorously evaluate our approach, we also present XThreatBench, a specialized multilingual dataset capturing fine-grained harmful behaviors drawn from real policy guidelines. Experiments with leading open-source LLMs (e.g., Llama, Qwen, Mistral) show that Soteria consistently improves safety metrics across high-, mid-, and low-resource languages. These findings highlight a promising path toward scalable, linguistically attuned, and ethically aligned LLMs worldwide.
Code (1)
Tasks
Safety AlignmentSimilar Papers 제목 키워드 기반
SOTERIA: In Search of Efficient Neural Networks for Private Inference
ML-as-a-service is gaining popularity where a cloud server hosts a trained model and offers prediction (inference) service to users. In this setting, our objective is to protect the confidentiality of both the users' inp…
Neural Architecture SearchFrom Weights to Activations: Is Steering the Next Frontier of Adaptation?
Post-training adaptation of language models is commonly achieved through parameter updates or input-based methods such as fine-tuning, parameter-efficient adaptation, and prompting. In parallel, a growing body of work mo…
Mechanistic Finetuning of Vision-Language-Action Models via Few-Shot Demonstrations
Vision-Language Action (VLAs) models promise to extend the remarkable success of vision-language models (VLMs) to robotics. Yet, unlike VLMs in the vision-language domain, VLAs for robotics require finetuning to contend …
Soteria: Provable Defense Against Privacy Leakage in Federated Learning From Representation Perspective
Federated learning (FL) is a popular distributed learning framework that can reduce privacy risks by not explicitly sharing private data. However, recent works have demonstrated that sharing model updates makes FL vu…
Federated LearningInference AttackFBHM: Functional Benchmarking and Steering of VLMs for Hateful Meme Detection
Hateful meme detection remains a formidable challenge for vision-language models, as existing benchmarks are structurally observational - confounding rhetorical hate mechanisms with target community features and preventi…
Multimodal Reasoning