arXiv AI
Jun 2
Low-Resource Safety Failures Are Action Failures, Not Representation Failures
arXiv:2606. 01196v1 Announce Type: cross Abstract: Safety alignment learned in high-resource languages transfers poorly to low-resource languages.
Hard Negatives Reveal What Easy Negatives Hide: Cross-Lingual Harmfulness Representations Degrade with Resource Tier Under Hard Negatives
The study investigates how safety alignment in large language models, trained mainly in English, transfers to other languages. While models show near-perfect harmfulness detection (AUROC > 0.98) using unrelated harmless prompts (easy negatives), performance drops sharply in low‑resource languages when using surface‑similar benign prompts (hard negatives). This degradation persists across multiple languages and models, indicating that easy‑negative evaluation alone cannot confirm cross‑lingual harmfulness representation quality.
Related stories
arXiv AI
Aug 12
Why Do Safety Guardrails Degrade Across Languages?
arXiv:2605. 17173v2 Announce Type: replace-cross Abstract: Large language models exhibit safety degradation in non-English languages.
arXiv AI
Aug 20
Latent Space Refusal Anchoring for Low-Resource African Languages: Mechanistic Safety Recovery Without Retraining
The paper introduces Latent Space Refusal Anchoring (LSR‑Anchoring), a training‑free technique that extracts a refusal direction from English prompts and applies it to the residual stream of instruction‑tuned models at inference time. The primary variant, Mean‑Activation Steering (MAS), works across several architectures (Llama‑3‑8B, Llama‑3.1‑70B, Mistral‑7B‑Instruct, Qwen2.5‑7B), restoring safety for low‑resource African languages with minimal performance loss, while a refined SAE‑Derived Steering (SDS) further reduces KL divergence without degrading legitimate prompt performance. The method shows positive transfer for Yoruba, Igbo, Igala, and Hausa, but fails for Arabic, suggesting a geometric mismatch rather than a data scarcity issue.
Hugging Face Trending Papers
Aug 11
The Illusion of Cross-Lingual Safety in Low-Resource Languages
Safety alignment in large language models (LLMs) is largely developed in English, assuming these safeguards generalize across multilingual settings. However, this assumption remains underexplored and exposes a vulnerability in low-resource languages.
arXiv AI
Aug 11
Decided Upstream, Written Late: Locating and Pricing the Cross-Lingual Refusal Circuit of a Multilingual MoE
arXiv:2608. 08032v1 Announce Type: new Abstract: Safety alignment in multilingual models is uneven: a model that reliably refuses a harmful request in English will often comply with the same request in a lower-resource language.
arXiv AI
Aug 13
Localizing Safety Alignment: MLP Layers and Mid-Network Blocks Encode Refusal Behavior in Large Language Models
arXiv:2608. 11583v1 Announce Type: new Abstract: Safety alignment in large language models is often treated as a distributed property of the entire network, yet its practical brittleness suggests that refusal behavior may be concentrated in a smaller set of parameters.
arXiv AI
Jul 14
Minionese: Comprehensive Benchmark and Mechanistic Study of Multilingual LLM Safety
arXiv:2607. 10112v1 Announce Type: cross Abstract: Safety alignment in large language models remains brittle across languages: prompts reliably refused in English can elicit harmful compliance in non-English and low-resource settings.
arXiv AI
Jun 9
Sycophancy as a Multilingual Alignment Failure: How Safety Degrades Across Languages, Topics, and Models
arXiv:2606. 08451v1 Announce Type: cross Abstract: Safety-aligned large language models often exhibit sycophancy, which is the tendency to affirm users' opinions regardless of factual accuracy.
arXiv AI
Aug 5
Toward Understanding the Transferability of Adversarial Suffixes in Large Language Models
arXiv:2510. 22014v2 Announce Type: replace-cross Abstract: Discrete optimization-based jailbreaking attacks on large language models aim to generate short, nonsensical suffixes that, when appended onto input prompts, elicit disallowed content.
arXiv AI
Jun 30
The Heterogeneous Safety Impacts of Benign Multilingual Fine-Tuning
arXiv:2606. 28843v1 Announce Type: cross Abstract: Fine-tuning a large language model is a ubiquitous method for enhancing its capability on a specific downstream task.
arXiv AI
Sep 15
Semantic Fibers and Cross-Gram Interference: A Calculus of Safety Drift in Overcomplete Representations
arXiv:2609.14861v1 Announce Type: cross Abstract: A deployed language model may refuse a harmful request in English yet comply with its faithful translation, revealing a cross-lingual safety failure...
arXiv AI
Aug 17
SomaliBench Eval: Measuring English-to-Somali Refusal Gaps in Open-Weight Language Models
arXiv:2605. 25420v2 Announce Type: replace-cross Abstract: Large language model safety evaluation remains heavily English-centered, leaving low-resource languages under-measured even when models are deployed globally.