This study investigates whether large language models (LLMs) exhibit different safety vulnerabilities when prompted with emojis instead of plain text. By testing 50 emoji‑augmented prompts on four open‑source LLMs—Mistral 7B, Qwen 2 7B, Gemma 2 9B, and Llama 3 8B—the authors found varying success rates of unsafe responses: Gemma 2 9B and Mistral 7B each had a 10% success rate, Llama 3 8B had 6%, while Qwen 2 7B was fully resistant. A chi‑square test confirmed significant differences in the outcome distributions, suggesting that input representation can markedly affect model robustness.
By M P V S Gopinadh
The paper introduces RiskCueBench, a benchmark designed to separate intent recovery from safety behavior in large language models. It evaluates 25 model groups on 640 prompts covering 160 latent intents across eight risk categories, revealing that higher contextual specificity can reduce refusal rates and safety scores even after intent recovery. The study highlights that safety behavior may be miscalibrated to prompt surface form, underscoring the need for evaluations that condition on semantic recoverability.
By Yuwei Wang, Xingwang Liang, Yu Xie, Haotian Chi, Haijun Geng
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.
By Will Hawkins, Kaivalya Rawal, Jonathan Rystr{\o}m, Stratis Tsirtsis, Zihao Fu, Greta Warren, Ryan Brown, Eoin Delaney, Sandra Wachter, Brent Mittelstadt, Chris Russell
The paper introduces the ASCII Attack, a single‑turn, black‑box method that embeds a harmful request within ASCII art and presents it as artwork to a large language model. By framing the request as artistic critique, the model can provide operational details that a plain request would normally be refused. Experiments across eleven models and eight harm topics show that the attack succeeds in 62% of cases versus 42% for direct controls, with the most vulnerable model achieving a 93% success rate.
By Da Cheng Gu, Yifei Dong, Xinghao Yang, Yongshun Gong, Wei Liu
arXiv:2608.29589v1 Announce Type: new
Abstract: Text-to-image (T2I) safety guardrails fail to generalize equitably to non-standard dialects. Evaluating 23,080 paired prompts across five English diale...
By Minkyu Kim, Juhwan Choi, YoungBin Kim
arXiv:2608. 14577v1 Announce Type: cross Abstract: Frontier large language models (LLMs) safety evaluation has largely treated harmful generation as an attack outcome rather than as an object of analysis.
By Zhouyuan Ma, Yutao Wu, Hanxun Huang, Xiang Zheng, Xiao Liu, Yixin Cao, Zuxuan Wu, Xingjun Ma, Yu-Gang Jiang
arXiv:2607. 02079v1 Announce Type: cross Abstract: We present HaloGuard 1.
By Navaneeth Sangameswaran, Preetham S, Ashmiya Lenin
The paper introduces a framework to evaluate and diagnose the robustness of low‑resource multilingual text‑to‑speech systems when faced with complex text inputs such as numbers, dates, named entities, long sentences, code‑switched expressions, and punctuation structures. It assesses robustness across content consistency, language consistency, and generation stability, and proposes automatic metrics (character error rate, language ID accuracy, duration abnormal rate) along with a lightweight Text Risk Score (TRS) that predicts synthesis risk from interpretable text features. Experiments on Thai, Vietnamese, Swahili, and Indonesian TTS systems reveal distinct failure patterns and show that TRS correlates positively with content and duration errors, offering a low‑cost pre‑synthesis risk indicator.
By Tianlun Zuo, Ziyu Zhang, Tingzhi Mao, Zhonghua Fu, Lei Xie
arXiv:2605. 17173v2 Announce Type: replace-cross Abstract: Large language models exhibit safety degradation in non-English languages.
By Max Zhang, Ameen Patel, Sang T. Truong, Sanmi Koyejo
IndicSafeEval is a new evaluation framework that tests the safety robustness of large language models against persuasion-based jailbreak attacks in Indian languages. The benchmark includes 7,200 adversarial prompts covering ten safety-critical content categories, six persuasive strategies, and four languages (Hindi, Bengali, Marathi, Punjabi). Experiments show that model safety varies significantly across languages, prompt styles, and risk categories, revealing that current English-centric safety evaluations miss important multilingual vulnerabilities.
By Saikat Mondal, Mamta, Deeksha Varshney, Oana Cocarascu, Asif Ekbal
The paper critiques current encoded‑prompt safety benchmarks that focus only on harmful requests, showing that such tests can misrepresent a model’s safety. By evaluating the benign arm under the same encoding, the authors reveal a substantial drop in the harm gap—sometimes to zero—indicating that the encoding masks true safety deficiencies. Across multiple large models and training pipelines, they document that the encoding can either hide or falsely inflate safety metrics, and they identify twelve specific instrument defects that contribute to these misleading results.
By Haoyu Zhang, Haowen Xu, Xiao Luo, Hanwen Liu, Yang Chen, Zijian Xiao, Yi Feng, Xiangchen Guan, Mohammad Zandsalimy, Shanu Sushmita
arXiv:2606. 07706v1 Announce Type: cross Abstract: Vision-Language Models (VLMs) have demonstrated strong performance across multimodal tasks, yet their safety robustness remains an open challenge.
By Rishabh Makwana, Mamta, Deeksha Varshney, Oana Cocarascu