arXiv:2510. 10271v2 Announce Type: replace-cross Abstract: Unlike regular tokens derived from existing text corpora, special tokens are artificially created to annotate structured conversations during the fine-tuning process of Large Language Models (LLMs).
arXiv:2607. 00481v1 Announce Type: cross Abstract: Jailbreak attacks remain a critical threat to the safe deployment of large language models (LLMs).
arXiv:2609.05850v1 Announce Type: cross Abstract: Despite the significant efforts devoted to aligning large language models (LLMs) with human values and ensuring safe deployment, recent work has reve...
arXiv:2508. 10031v2 Announce Type: replace-cross Abstract: While Large Language Models (LLMs) have shown significant advancements in performance, various jailbreak attacks have posed growing safety and ethical risks.
The paper "Jailbreaking in the Haystack" introduces NINJA, a jailbreak technique that exploits long-context language models by appending benign, model-generated content to harmful user goals. It demonstrates that the position of harmful goals within the context is crucial for safety, and shows that NINJA significantly boosts attack success rates on models such as LLaMA, Qwen, Mistral, and Gemini. Unlike previous methods, NINJA is low-resource, transferable, less detectable, and compute‑optimal, revealing that carefully crafted benign long contexts can expose fundamental vulnerabilities in modern LMs.
The paper introduces Semantic Overlays, a steering technique that adds non‑textual annotations to a language model’s input by applying learned adapters at specific prefill positions. These overlays create an out‑of‑band channel that encodes span identity and complex semantics, enabling the model to interpret marked text differently—such as rewriting code in a specified language or ignoring executable instructions. Experiments show that Semantic Overlays dramatically reduce prompt‑injection success rates while preserving model utility and readability of marked spans.
arXiv:2609.26185v1 Announce Type: cross Abstract: Large Language Models (LLMs) demonstrate impressive capabilities across many applications but remain vulnerable to jailbreak attacks, which elicit ha...
arXiv:2605.27110v2 Announce Type: replace-cross Abstract: In this work, we propose BAIT (Boundary-Aware Iterative Trap), a three-step jailbreak framework that elicits malicious information through in...
Reflex-Guard is a lightweight, locally‑run guardrail for large language models that uses jailbreak‑aware preprocessing, compact sentence‑transformer embeddings, and seven fast binary classifiers to filter unsafe prompts. It achieves 95.9 % recall on harmful prompts with an end‑to‑end latency of 37.6 ms, outperforming existing solutions such as Llama Guard 2 (255 ms) and SafeDecoding (723 ms). The system can detect all GCG suffix attacks and Base64‑encoded prompts at the default threshold, while DrAttack structured prompts require a lower threshold of 0.03 for optimal detection, and it attains a Reflex Efficiency Score of up to 16.79.
arXiv:2510.17904v3 Announce Type: replace-cross Abstract: Large Language Models (LLMs) are widely used because they process structures, syntax and code well, but this same ability also makes them par...
Reflex-Guard is a lightweight, locally running guardrail for large language models that uses jailbreak-aware preprocessing, compact sentence‑transformer embeddings, and seven fast binary classifiers to filter unsafe prompts. It achieves 95.9% recall on harmful prompts with an end‑to‑end latency of 37.6 ms, far faster than existing solutions such as Llama Guard 2 (255 ms) and SafeDecoding (723 ms). The system can detect all GCG suffix attacks and Base64‑encoded prompts at the default threshold, and it attains a Reflex Efficiency Score up to 16.79, outperforming its competitors.
arXiv:2609.06540v1 Announce Type: cross Abstract: Large language models (LLMs) are increasingly deployed in safety-critical applications, yet jailbreak attacks can conceal harmful intent through role...
The paper examines how open‑weight language models expose the control tokens used in chat templates, allowing attackers to forge turn boundaries that the model treats as legitimate. An audit of 256 deployed tokenizers shows all are vulnerable, and the commonly recommended flag fails to protect 56.6% of cases. The authors introduce nameless tokenization, which removes surface strings for control identifiers while preserving their internal representation, achieving identical token streams on clean data and significantly improving accuracy on delimiter‑bearing text.