The paper investigates safety risks in model merging, showing that even when all constituent models are individually safety‑aligned, merging can expose a jailbreak vulnerability rooted in the pretrained foundation model. It introduces Basin‑Aware Jailbreak (BAJ), a min–max optimization method that generates adversarial suffixes transferable across merged models sharing the same backbone, without needing the exact merging coefficients or checkpoints. Experiments demonstrate BAJ’s high transfer success rates across diverse backbones and merging settings, and its resilience against existing defenses.
arXiv:2512. 14751v3 Announce Type: replace-cross Abstract: Finetuning pretrained large language models (LLMs) has become the standard paradigm for developing downstream applications.
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.
arXiv:2506. 22666v3 Announce Type: replace-cross Abstract: The rise of API-only access to state-of-the-art LLMs highlights the need for effective black-box jailbreak methods to identify model vulnerabilities in real-world settings.
arXiv:2606. 16242v1 Announce Type: new Abstract: The Rapid Response (RR) framework, deployed in production systems, including Anthropic's ASL-3 safeguards, continuously improves jailbreak-detection classifiers.
arXiv:2602. 12418v2 Announce Type: replace-cross Abstract: Jailbreak attacks remain a persistent threat to large language model safety.
arXiv:2607. 19424v1 Announce Type: cross Abstract: The assessment of jailbreak attacks against large language models currently suffers from inconsistent evaluation criteria and methods, leading to unreliable estimates of attack success rates.
arXiv:2502. 09755v4 Announce Type: replace-cross Abstract: Safety-aligned LLMs respond to prompts with either compliance or refusal, each corresponding to distinct directions in the model's activation space.
arXiv:2609.02172v1 Announce Type: cross Abstract: Optimization-based jailbreak attacks such as Greedy Coordinate Gradient (GCG) achieve strong effectiveness and transferability by optimizing adversar...
arXiv:2608. 03070v1 Announce Type: cross Abstract: Frontier AI model developers increasingly rely on layered safeguards to prevent catastrophic misuse, but little public evidence exists on how much protection these safeguards provide, or how consistently across developers.
arXiv:2606. 28153v1 Announce Type: cross Abstract: Jailbreak attacks bypass LLM safety alignment, yet their mechanisms remain poorly understood.
arXiv:2509. 03985v2 Announce Type: replace-cross Abstract: In deployment and application, large language models (LLMs) typically undergo safety alignment to prevent illegal and unethical outputs.
arXiv:2508. 10029v3 Announce Type: replace-cross Abstract: Safety-aligned large language models can still be manipulated through white-box interventions that modify their internal representations.