A neuron-level cross-dimensional safety alignment framework driven by modality- and language-shared safety neurons (MLS-Neurons) that significantly outperforms state-of-the-art approaches across diverse multilingual and multimodal safety benchmarks while preserving general utility.
Abstract
As large vision-language models (LVLMs) are deployed globally, the combination of multilingual instructions and visual information makes malicious attacks more covert and sophisticated than ever before. However, existing methods isolate language and modality defenses, which, coupled with the scarcity of safety data and high fine-tuning costs, makes it difficult for models to defend against compound attacks. To address this severe challenge, we propose a neuron-level cross-dimensional safety alignment framework driven by modality- and language-shared safety neurons (MLS-Neurons). First, we identify monolingual and unimodal safety neurons by comparing responses to harmful and benign samples, quantifying functional saliency through activation strength and downstream impact. Then, by intersecting these unimodal neurons within each language, we extract modality-shared safety neurons (MS-Neurons) responsive to both visual and textual risks, bridging the safety representation gap between modalities. Furthermore, using English as a semantic anchor, we intersect MS-Neurons across languages to identify modality- and language-shared safety neurons (MLS-Neurons), serving as key defenses against compound attacks. Finally, we update only this minimal subset of shared neurons (~0.03% of parameters), transferring English-only safety supervision to multilingual and multimodal scenarios. Extensive experiments show that our method significantly outperforms state-of-the-art approaches across diverse multilingual and multimodal safety benchmarks while preserving general utility.
Through rigorous mechanistic analysis, this work identifies the Ghost Anchor phenomenon: a temporal modality asynchrony where linguistic translation to the English semantic manifold completes in early layers, while visual semanticization remains immature.
Yihang Du, Juhao Liang, Zheng-Zhao Lai et al.· 0 citations
A multilingual medical VQA benchmark over eight languages is constructed, organized into four representative scenarios that isolate the core capabilities medical VQA requires, and a training-free scenario-aware representation engineering method is proposed, leveraging LVLMs's superior English medical VQA capability to steer non-English representations toward their English counterparts at inference time.
Jingbo Wang, Sendong Zhao, Haochun Wang et al.· 0 citations
The findings suggest that activation steering may provide a practical, low- cost mechanism for extending English-derived safety signals to other languages, and introduce a multilingual translation-and-evaluation pipeline to facilitate future work on cross-lingual safety interventions.
Emma V. Stein, Dominik Meier, Terry Ruas et al.· 0 citations
Although Large Language Models (LLMs) have demonstrated promising safety performance, extending them to Multimodal Large Language Models (MLLMs) exposes a significant gap between expanded multimodal capabilities and existing safety mechanisms. Current defenses remain predominantly confined to specific modal settings, thereby limiting their robustness against broader cross-modal threats. To bridge this gap, we introduce SafeNexus, a cross-modal safety alignment framework that adopts a dedicated neuron-level intervention strategy. First, we formulate a neuron localization paradigm that identifies functionally specialized neurons by characterizing intermediate-layer activation patterns and quantifying their functional salience through importance scoring. Building upon this paradigm, we exploit contrastive data to identify modality-bound safety neurons (BS-Neurons), and validate their role in regulating safety behavior within each modality via targeted suppression. Further cross-modal analysis defines modality-universal safety neurons (US-Neurons) as the shared subset of BS-Neurons identified across individual modalities, serving as the core for defending against harmful cross-modal attacks. We observe that suppressing these neurons substantially degrades safety performance across modalities, while leaving overall utility largely unaffected. Building on these insights, we propose two safety alignment strategies: activation-level safety amplifier and safety neuron calibrator. The proposed strategies enhance model safety through two distinct routes: the former amplifies the activation magnitudes of US-Neurons, while the latter selectively calibrates them via targeted fine-tuning. Extensive experiments demonstrate that our method outperforms prevailing state-of-the-art approaches on safety benchmarks spanning diverse modality combinations, while effectively preserving utility.
A Low-Complexity Cross-Modal Alignment via Projection (LCAP) network is proposed, which introduces Projective Token Compression (PTC), which leverages Mish activation and adaptive average pooling to reduce feature redundancy while enhancing discriminative information, and Positional Spatial Enhancement (PSE), which explicitly injects positional cues into the compressed representations and strengthens spatial structure.
Yuchen Sha, Lingli Wan, Ge Yang et al.· The Visual Computer· 0 citations
Experiments show that MMAligner raises the average refusal rate on unsafe multimodal inputs to 99% with less than 2% utility degradation and minimal training data, substantially improving the safety-utility trade-off over existing baselines.
Shenyi Zhang, Keyan Guo, Zihao Wang et al.· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
