Jailbreak Evaluations

LLM supervision systems, namely input/output moderation filters and jailbreak detectors, are the primary safeguard against misuse in deployed AI applications, yet existing benchmarks are often vendor-biased, omit cost and latency, and rarely compare specialized guardrails against repurposed generalist LLMs. We present BELLS-O (Benchmark for the Evaluation of LLM Supervision Systems, Operational), the first independent operational benchmark of LLM supervision systems. BELLS-O evaluates 28 systems from 17 providers: every major specialized guardrail (e.g., LlamaGuard-4, ShieldGemma-2, Lakera Guard) and frontier generalists repurposed as supervisors (e.g., GPT-5.4, Claude Sonnet 4.6, Grok-4.1), jointly on detection rate, false-positive rate, latency, and monetary cost. We cover input/output moderation across 11 harm categories and jailbreak detection across 13 attack techniques, using in-house datasets built from handcrafted prompts, expert-curated samples, and quality-controlled synthetic generation. To suppress latent generator fingerprints in synthetic data, every generated sample is paraphrased. Mapping the Pareto frontier reveals use-case-dependent tradeoffs. On content moderation, specialized supervisors are operationally dominant: top systems match frontier LLMs on detection (~95% vs. 94%) at comparably low false-positive rates (<=2%), while running 5-10x faster and ~10x cheaper. On jailbreak detection, the tradeoff shifts: frontier LLMs achieve higher detection and lower false-positive rates but at 10-50x higher cost and 5-10x higher latency. We release the benchmark, framework, leaderboard, and datasets as the first vendor-neutral basis for selecting safeguards under real deployment constraints.

Stream guardrails enable token-level safety detection before full responses are generated. However, they often make overly conservative judgements and block those sensitive but safe tokens, which is known as over-refusal. Due to lack of full context, they also fail to detect implicitly harmful content from jailbreaking. To address these challenges, we propose FreoStream, a novel streaming guardrail framework. Specifically, FreoStream fine-tunes a LoRA module to perform Future-Aware Reasoning when the base guardrail detects unsafe tokens. The reasoning process follows a Future-Reason-Judge paradigm: predict the future, reason about the full context and give the final judgement. This design can effectively reduce over-refusal by incorporating the future information. Moreover, we introduce the Safety-Aligned Optimization module that extracts the safety-aligned component from the reasoning gradients to update the base guardrail model, thereby enhancing streaming safety detection. Extensive experiments on various safety benchmarks demonstrate that FreoStream achieves lower over-refusal rates and better jailbreak defense compared to existing streaming guardrails.

Large Language Models (LLMs) are increasingly used for code generation, raising concerns that they may be misused to produce malicious code. Meanwhile, Grammar-Constrained Decoding (GCD) has been widely adopted to improve the reliability of LLM-generated code by enforcing syntactic validity. In this paper, we reveal a counterintuitive risk: this reliability-oriented technique can itself become an attack surface. We uncover a new jailbreak attack, termed CodeSpear, that exploits GCD to induce LLMs into generating malicious code. Our experiments show that simply applying a benign code grammar constraint can effectively jailbreak LLMs. To address this vulnerability, we propose CodeShield, a safety alignment approach that robustly preserves safe behavior even under attacker-controlled grammar constraints. CodeShield aligns the model in the code modality by teaching it to generate honeypot code under GCD. Such code is semantically harmless, so it does not implement the malicious request, and structurally diverse, so it is difficult to suppress through grammar tightening. At the same time, CodeShield still preserves natural-language refusals when natural language is available. Experiments on 10 popular LLMs across 4 benchmarks show that CodeSpear outperforms representative jailbreak baselines and increases the attack success rate by more than 30 percentage points on average. CodeShield also restores safety under CodeSpear while preserving benign utility. Our findings reveal a fundamental risk of GCD and call for greater attention to its potential security implications.

Jailbreak attacks expose persistent safety weaknesses in large language models (LLMs), but existing stateless single-turn methods face a trade-off: hand-crafted prompts are expressive but static, while iterative prompt optimization can adapt but often relies on low-level mutations that require many target queries. We propose JailbreakOPT, a tool-assisted framework for improving iterative single-turn jailbreak prompt optimization. JailbreakOPT organizes diverse atomic jailbreak prompts into an attack tool library and composes them through a unified intra-episode optimization abstraction to generate stronger standalone attack prompts. To reuse experience across attack episodes, JailbreakOPT further frames tool selection as a contextual bandit problem and applies contextual Thompson sampling to guide exploration and exploitation based on past outcomes. Experiments across multiple target LLMs and attack goals show that JailbreakOPT improves attack success rate (ASR) while reducing the number of attacks until success (No.A) compared with atomic single-turn attacks and existing iterative optimization baselines. This paper may contain offensive or harmful content.

Tool-using LLM agents interact with the world through actions that persist state in artifacts (e.g., workspace files or logs). Consequently, jailbreak defenses must reason about cross-step composition rather than isolated text. Yet most existing attacks and defenses, including ``multi-turn'' jailbreaks such as Crescendo and Tree of Attacks,still assume a single contiguous conversation visible to the defender. This assumption breaks down in real agent pipelines, where enforcement is fragmented across tools, modules, and time, and where artifact provenance is often not tracked. We operationalize a deployment failure mode for tool-using LLM agents, the \emph{provenance gap}, and study reproducible triggers for it: \emph{Context-Fractured Decomposition} (CFD), a family of cross-context multi-step jailbreaks that preserve benign-looking intermediate artifacts from an early interaction and elicit harmful behavior much later, potentially in a different agent instance or workflow stage, via individually innocuous tool actions whose risk emerges only under delayed artifact-mediated composition. We instrument the failure mode with trace-level diagnostics and outline a verifiable mitigation direction (provenance lineage tagging). Across agent-system jailbreak benchmarks, CFD improves success rates by up to 28.3 percentage points over state-of-the-art baselines, even against strong single-turn judges. Disclaimer: This paper contains examples of harmful or offensive language.

Vision-Language Models (VLMs) have demonstrated strong performance across multimodal tasks, yet their safety robustness remains an open challenge. While prior work has shown that structured visual prompts such as flowcharts can effectively jailbreak VLMs, existing studies are largely limited to English-centric settings. In this paper, we introduce MLingualFC, a multilingual multimodal benchmark designed to evaluate jailbreak vulnerabilities of VLMs across diverse languages using structured flowchart representations. MLingualFC encodes harmful instructions into flowchart images across five languages (Hindi, Punjabi, Spanish, Romanian, and German). We evaluate state-of-the-art multilingual VLMs, including Qwen2.5-VL, Gemma-4, and Pangea, under a black-box threat model. Our results reveal significant multilingual safety gaps. Flowchart-based attacks achieve high attack success rates (ASR) in case of Latin script languages, demonstrating that visual encoding of harmful content effectively bypasses safety alignment across languages. In contrast, non-Latin script languages such as Punjabi exhibit substantially lower ASR, suggesting potential limitations in visual text recognition rather than stronger safety alignment. These findings highlight that current VLM safety mechanisms fail to generalize across languages and modalities. Resources are available at https://github.com/Rishabhpm23/MLingualFC

Large audio language models (LALMs) are increasingly deployed in real-world applications, yet their safety alignment is still primarily evaluated on monolingual, text-based harmful prompts. This leaves their generalizability under multilingual and spoken settings, particularly code-switched speech, largely underexplored. To address this gap, we introduce SpeechJBB, an audio jailbreak dataset for benchmarking across multiple state-of-the-art LALMs. The extent of safety weaknesses is further probed by introducing an augmented setting where phonologically plausible pseudo-words are inserted around safety-critical terms to simulate localized obfuscation. Across models, code-switched harmful audio yields substantially high jailbreak success rates (JSR), with non-English monolingual and non-English code-switched pairs exhibiting the highest attack success. Pseudo-word insertion further reduces refusal rates, which demonstrates that natural-sounding obfuscation can effectively bypass safety policies.

Despite advances in safety alignment, large language models remain vulnerable to continuously evolving jailbreaks. Existing fine-tuned safety classifiers cannot adapt to these evolving attacks, while adaptive memory-based guardrails tend to over-refuse benign queries that resemble stored attacks. We propose Membrane, a self-evolving guardrail built on Contrastive Safety Memory (CSM): each cell pairs the conditions for blocking a harmful query with those for permitting a superficially similar benign request. Without retraining, Membrane evolves CSM by distilling each harmful interaction and its benign counterpart into a contrastive cell indexed by the underlying attack strategy, so that one cell generalizes across topical variants of the same mechanism. At inference, retrieved cells serve as grounding context for precise safety decisions. Across model-level safety on HarmBench and agent-level safety on AgentHarm, Membrane achieves the highest F1 on all six jailbreak attacks. Notably, benign refusal on AgentHarm stays at 7-14%, well below the 28-85% range of prior guards. Memory cells also retain 87-88% F1 under cross-attack transfer and remain stable under memory poisoning.

Large language models (LLMs) are rigorously aligned to refuse harmful requests, a process that inherently cultivates a latent capacity to evaluate and recognize unsafe content. In this work, we reveal that this advanced safety awareness inadvertently introduces a fatal vulnerability. We introduce Posterior Attack, a single-query jailbreak that bypasses guardrails by prompting the model to generate the exact harmful response its internal classifier would normally flag as unsafe. Through extensive empirical evaluation across 30 open-source LLMs (up to 35B parameters in size) and frontier models (e.g., GPT-5, Claude 4.6), we observe a striking phenomenon: models with superior safety-judgment capabilities are disproportionately more susceptible to this exploitation. To explain this, we formalize the Safety Paradox, analytically showing that monotonic improvements in safety alignment naturally amplify posterior vulnerability. Finally, we establish a causal link via reinforcement learning interventions, exemplifying that artificially degrading a model's safety judgment immunizes it against the attack, whereas enhancing judgment exacerbates the vulnerability. Our findings highlight potential flaws in current alignment paradigms, indicating that defense mechanisms may require further structural refinement.

As large language models (LLMs) are widely deployed, identifying their vulnerability through jailbreak attacks becomes increasingly critical. Optimization-based attacks like Greedy Coordinate Gradient (GCG) have focused on inserting adversarial tokens to the end of prompts. However, GCG restricts adversarial tokens to a fixed insertion point (typically the prompt suffix), leaving the effect of inserting tokens at other positions unexplored. In this paper, we empirically investigate \emph{slots}, i.e., candidate positions within a prompt where tokens can be inserted. We find that vulnerability to jailbreaking is highly related to the selection of the \emph{slots}. Based on these findings, we introduce the \textit{Vulnerable Slot Score} (VSS) to quantify the positional vulnerability to jailbreaking. We then propose SlotGCG, which evaluates all slots with VSS, selects the most vulnerable slots for insertion, and runs a targeted optimization attack at those slots. Our approach provides a position-search mechanism that is attack-agnostic and can be plugged into any optimization-based attack, adding only 200ms of preprocessing time. Experiments across multiple models demonstrate that SlotGCG significantly outperforms existing methods. Specifically, it achieves 14% higher Attack Success Rates (ASR) over GCG-based attacks, converges faster, and shows superior robustness against defense methods with 42% higher ASR than baseline approaches. Our implementation is available at \href{https://github.com/youai058/SlotGCG}{https://github.com/youai058/SlotGCG}

Large Language Models (LLMs) have transformed natural language processing, but they remain vulnerable to Prompt Injection (PI) and Jailbreak (JB) attacks. In addition, benchmark evaluations may be affected by contamination and partial information leakage, compromising performance estimates. This work presents GuardNet, a guardrail system based on an ensemble of shallow neural networks (BiLSTMs) with approximately 47 million parameters. We investigate the hypothesis that robustness in adversarial scenarios depends more on the diversity of example coverage and threshold calibration than on model scale. The results indicate that GuardNet achieves competitive performance compared with lightweight detectors and high efficiency at low latency, although larger LLMs such as Mistral-7B and Llama-3.1-8B still achieve superior performance in terms of F1 score and AUROC on the blind JBB-Behaviors benchmark. Nevertheless, GuardNet achieves an AUROC of 0.747 on the blind dataset (n = 200) and an F1 score of 0.92 on a proprietary benchmark (n = 50), under threshold calibration and evaluation with declared partial information leakage. The system operates with an average latency of approximately 50 ms on CPU, making it suitable for deployment in production environments with cost and infrastructure constraints.

Existing jailbreaks against aligned LLMs are discrete artifacts whose surface forms are easy to fingerprint and patch. We argue that the broader failure mode may lie not in any specific prompt, but in natural writing registers that safety tuning under-covers. Building on this insight, we introduce VAR, the first jailbreak family that uses real fanfiction subgenres as universal attack carriers: a creative-writing meta prompt is conditioned on passages from one of twelve Archive of Our Own (AO3) subgenres, and the harmful behavior is embedded as the climax of the resulting scene. The construction requires neither an adversarial attacker LLM nor optimization. On eight aligned LLMs over the union of HarmBench and JailbreakBench, this attack lifts mean ASR from 0.278 to 0.731 under a four-judge ensemble; a factorial decomposition shows the gain is carried by register rather than length or structure. Two active defenses widen rather than narrow the vernacular-to-baseline ratio, indicating that template-targeting defenses merely steer attackers toward register-based attacks like ours. We also propose VAR-A4, a static four-turn extension that attains a mean ASR of 0.924, substantially exceeding three existing multi-turn methods. Our code and data are safely open-sourced at https://github.com/T-Lab-CUHKSZ/VAR.

Accurately evaluating adversarial robustness is a longstanding challenge. A flawed attack design can inflate robustness estimates, making deployment risk assessment and defense comparison unreliable. Historically, standardized attacks such as AutoAttack have largely resolved this for image classifiers, providing a reliable evaluation baseline for systematic comparison across defenses. However, no equivalent exists for LLM jailbreak evaluation yet, where designing such an attack is considerably more difficult. A reliable attack must, among other things, be black-box compatible, applicable to arbitrary defense pipelines, and efficient, which no existing method jointly satisfies. We introduce Indirect Harm Optimization (IHO), a masked diffusion language model attacker trained via iterative preference optimization against a harmfulness judge, requiring only black-box access to the target. The same method can be used without modification as a strong adaptive attack on individual behaviors, or as an efficient amortized policy that transfers to held-out behaviors and unseen target models without fine-tuning. Even against layered defenses, such as a Circuit Breaker-trained model combined with an auxiliary detector, IHO improves attack success considerably over state-of-the-art approaches, without any defense-specific adaptation. Our results position IHO as a practical step toward the kind of standardized jailbreak evaluation that has improved reliability in the past. Code and models are available on GitHub and Hugging Face.

Multi-turn jailbreak attacks on large language models (LLMs) reveal a mismatch in current guardrails: they operate on individual turns, while attacks unfold as trajectories across conversations. We propose a shift from content to dynamics, modeling conversations as paths in representation space and asking whether adversarial intent is encoded early in their geometry. We introduce PsychoPass, a framework that extracts geometric features from conversation trajectories in embedding space to predict a potential attack before harmful content is produced. These features achieve near-perfect performance in naïve classifiers, which is largely explained by the inclusion of number of turns as a feature. After removing this confound, a smaller but consistent geometric signal remains, with classification performance that does not depend meaningfully on encoder choice. Crucially, this signal appears early in the conversation: attack outcomes remain above chance from short prefixes alone, more reliably than baseline guardrails. A supporting theoretical analysis explains these findings via a decomposition of length and shape, a detection bound based on prefix length, and encoder invariance. Together, these results show that adversarial conversations leave an early, representation-robust geometric fingerprint suitable for online monitoring.

Large language models remain vulnerable to jailbreak backdoor attacks, where adversaries poison safety alignment data to embed hidden triggers that bypass safety mechanisms. Existing defenses often require comprehensive attack information or multiple triggered examples, making them impractical when defenders only observe a single reported failure case without knowing whether it stems from a backdoor attack or a natural alignment bug. This paper presents Patcher, a post-hoc defense framework that repairs backdoored language models using only a single reported failure case and the model parameters. Patcher operates in two stages. First, it localizes backdoor triggers by computing response-conditioned gradient-based saliency scores and applying adaptive clustering to separate triggers from benign context. Second, it patches the model through a constrained fine-tuning objective that breaks the trigger-response association while preserving benign-task utility and robustness to non-triggered jailbreak attacks through KL-divergence constraints. We conduct extensive evaluations across multiple backdoor attack strategies and demonstrate that Patcher successfully localizes triggers and neutralizes backdoors while maintaining model utility. We further show robustness against adaptive attacks designed to evade our defense. This work represents a significant step toward practical defenses against training-time attacks in deployed language models.

Diffusion large language models (dLLMs) generate text by iteratively denoising partially masked sequences under bidirectional context, exposing a safety surface distinct from autoregressive LLMs. Because mask tokens are native inputs and tokens are committed by confidence rather than position, harmful content can be induced through infilling and outside the monitored prefix. Existing jailbreaks either miss this native infill capability or rely on low-diversity mask-bearing templates applied uniformly across goals, with little structural adaptation or accumulated attack experience. We propose MaskForge, a fully black-box adaptive attack that casts dLLM red-teaming as optimized search over a growing library of structural patterns. MaskForge abstracts successful attempts into reusable schemas, selects goal-compatible patterns with a UCB bandit, and invokes a scorer-guided fallback when the current library fails. Successful attempts are distilled back into the pattern library, enabling experience to accumulate across goals. Across five public dLLMs and three benchmarks, MaskForge achieves an average attack success rate of 79.3%, a 17.6% relative improvement over the strongest competing dLLM baseline. The matured pattern library further transfers to AdvBench without any updates, achieving a 88.2% attack success rate and a 67% relative improvement over the strongest competing baseline.

As multimodal large language models (MLLMs) have advanced to process video inputs, concerns have emerged about their potential for malicious misuse. Prior jailbreak studies have shown that safety alignment in MLLMs can be bypassed through visual inputs, yet it remains unclear which properties of video inputs induce this vulnerability. To address this gap, we introduce Multi-Clip Video (MCV) SafetyBench, a dataset of 2,920 videos designed to evaluate how the diversity of video inputs affects the vulnerability of MLLMs. Each video consists of multiple short clips depicting diverse contexts related to a harmful query. Experiments on eight representative video MLLMs show that attack success consistently increases with the number of clips. Our results further indicate that the video modality is (1) more vulnerable than the image modality, (2) more vulnerable to dynamic videos than to static videos, and (3) more vulnerable when videos contain more diverse contexts. Building on these findings, we propose a defense strategy that leverages the relative robustness of the image modality.

Multi-turn jailbreak attacks pose a growing threat to LLMs by exploiting conversational dynamics such as gradual escalation and cross-turn coordination. Existing defenses either rely on costly retraining -- often degrading model utility -- or apply single-turn analysis independently at each turn, failing to capture how risk accumulates along interaction trajectories. We observe that safety behavior in multi-turn interaction is trajectory-dependent: dialogue history continuously reshapes the model's conditioning context, making it insufficient to evaluate each turn in isolation. Motivated by this insight, we present THRD, the first training-free framework that explicitly models temporal risk accumulation for multi-turn jailbreak defense. THRD integrates four modules: a Turn-level Risk Assessor (TRA) for instantaneous risk estimation, a Historical Context Analyzer (HCA) for cross-turn intent escalation detection, a Response Evaluator (RE) for identifying facilitative outputs, and a Decision Module that combines these signals through a time-evolving scoring mechanism with attenuation-based modulation and trend-aware adjustment. Experiments against state-of-the-art multi-turn attacks -- including tree-search-based and multi-agent collaborative methods -- across two target models show that THRD reduces ASR to 0.2--4.0% while preserving model utility within 1.5% degradation on MMLU and GSM8K. Ablation studies confirm non-redundant module contributions and stable cross-architecture generalization. Analysis of first rejection triggers reveals that over 70% of multi-turn attacks require Turn~2 or later to detect, validating the necessity of explicit temporal aggregation.

Safety evaluation of Large Language Models (LLMs) remains heavily English-centric, leaving Low-Resource Languages (LRLs), particularly African ones, critically underexplored. We introduce TUKABENCH, a jailbreak benchmark for seven African languages that extends JailbreakBench (JBB) beyond direct translation through four settings: human translation of JBB prompts, English adaptation to African contexts followed by human translation, human-curated prompts validated through interactions with GPT-5.2, and code-switched prompts combining English and African languages, isolating the effect of language, cultural grounding, and prompt evasiveness on model safety. Across closed and open models, prompting in African languages reduces refusal relative to English, with culturally adapted prompts leading to least refusal. The evaluation also surfaces two structural limitations: model comprehension failures and reduced LLM-as-a-judge reliability in LRLs. To capture the first, we introduce Deflection alongside Refused and Jailbroken; to assess the second, we validate outputs with human annotations, showing that judge-human agreement drops in lower-resource languages and less commonly supported scripts.

Multi-turn jailbreak attacks pose a growing threat to large language model (LLM) safety because they exploit feedback from auxiliary judge models to iteratively refine prompts toward harmful goals. Existing defenses largely detect or block unsafe content at individual turns or at the final response, leaving the judge-driven refinement loop intact and allowing attackers to extract informative feedback from intermediate interactions. We introduce D-Judge, a semantics-preserving output rewriting defense that intervenes directly in this loop by rewriting the victim LLM's responses before they are evaluated by the attacker's judge. By misaligning the judge's feedback signal without changing the meaning of the original response, D-Judge derails the attacker's prompt-refinement process, causing subsequent queries to be optimized against a distorted signal of attack progress. To improve D-Judge's ability to produce such rewrites, we construct a dataset of semantically equivalent response pairs that induce different judge-assigned harmfulness scores, and use it for supervised fine-tuning followed by direct preference optimization. Experiments on HarmBench show that D-Judge reduces the success rate of state-of-the-art multi-turn jailbreaks while preserving performance on benign benchmarks.

Patient-facing medical chatbots are commonly evaluated on single-turn prompts, yet real users push back after refusals, add urgency, and invoke authority. We introduce MultiTurnPSB, a four-turn adversarial extension of PatientSafetyBench, and evaluate GPT-4.1-mini under fixed template, template-adaptive, and live adversarial attacks. Unsafe responses rise from 35% to nearly 80% by Turn 4 under live attack. Under the same adversary, GPT-4.1-mini and Claude Sonnet 4.5 are statistically indistinguishable at baseline but diverge to a 19x gap by Turn 4, a difference invisible to single-turn evaluation. We characterize four degradation trajectory signatures and identify a two-element attack formula responsible for most catastrophic failures. A lightweight input-side classifier reduces Turn 4 unsafe responses by 52 percentage points despite severe accuracy degradation, but the 45% false alarm rate on benign queries is the primary deployment constraint. A methodological finding also emerges: Claude Sonnet refused to generate adversarial messages in over half of late-turn conversations despite explicit red team framing, suggesting safety training may generalize to the attacker role.

As Large Language Models evolve for user convenience, vulnerability to jailbreak attacks continues to be reported despite ongoing efforts in safety training. Traditional jailbreak techniques typically focus on a single prompt injection, neglecting the models' ability to remember the flow of conversation and the user's instructions. In this paper, we propose Persona Attack, a memory injection based jailbreak method that manipulates the model's context window through a step by step approach. Experimental results from applying Persona Attack to several widely used LLMs reveal that, as injections accumulate in memory, models increasingly prioritize these instructions over their internal safety alignment mechanisms. Furthermore, our experiments empirically demonstrate that the attack success rate varies not only according to the memory implementation of the model, but also combinations of instructions and can reach 95% under specific instruction configurations.

Large Audio Language Models (LALMs) expand jailbreak risks from token-level prompting to the full speech perception-to-reasoning pipeline, where unsafe behavior can be induced through semantics, acoustic style, signal artifacts, or internal representations. Existing work studies these risks under heterogeneous threat models and evaluation protocols, making it difficult to compare attack practicality or defense utility. This paper provides a unified taxonomy and a controlled empirical evaluation of LALM jailbreak attacks and defenses. We organize prior work into semantic, acoustic, signal, and embedding-layer attacks; guard-based, training-free, and training-based defenses; and cross-modal, audio-native, and interactive benchmarks. We then evaluate representative attacks and defenses across ten open-source LALMs, measuring not only attack success rate but also benign refusal and latency. Our results show that Acoustic Best-of-N reveals strong worst-case audio-space vulnerabilities, Narrative Framing is an effective low-latency semantic threat, and current defenses trade robustness against benign usability. These findings support cost- and utility-aware evaluation as a necessary complement to success-rate-only LALM safety benchmarks.

Real-time safety filtering for large language model (LLM) applications requires classifiers that can detect unsafe prompts, toxic language, jailbreak attempts, and unsafe responses without the cost profile of large guardrail models, and that can distinguish benign sensitive text from genuinely covert harmful content. In this paper, we introduce Opir, a family of encoder-based guardrail models built on the GLiClass architecture. Opir includes multi-task models for binary safe/unsafe classification, multi-label toxicity classification, jailbreak classification, and zero-shot unsafe prompt and response categorization. We also release edge variants with fewer than 100M parameters dedicated to binary safe/unsafe categorization. The models are trained on a three-level taxonomy containing 996 categories across 16 top-level labels, 126 mid-level labels, and 854 leaf labels. Opir's training data combines taxonomy-grounded unsafe prompts, adversarially mined hard negatives, benign safety-preserving examples, generated response examples, multilingual translations, and portions of the Aegis2 and WildGuard training subsets. We also open-sourced an evaluation harness that supports GLiClass and GLiNER2 backends as well as decoder-based models, and covers binary safety classification, multi-label categorization, toxicity, jailbreak detection, prompt safety, response safety, response refusal, and prompt subcategory views across public benchmark families. Across an expanded comparison spanning 12 safety-classification tasks and 17 category tasks against eight contemporary guardrail systems -- including both GLiNER2-based and generative guardrail models -- Opir variants are competitive on or ahead of the strongest open-weight baselines on the majority of benchmark datasets while operating with a substantially smaller deployment footprint.

Attack Success Rate (ASR) evaluates each jailbreak with a single yes/no label at the end of generation, telling us whether a failure happened but not how it unfolded. Two attacks that produce equally harmful outputs may have followed completely different paths, and ASR cannot tell them apart. We make those hidden paths observable from logits alone. Temporal Logit Observability (TLO) is a training-free diagnostic that watches a compliance-refusal margin during decoding and places each model-attack condition on a calibrated 2D plane. By design, this plane is most informative exactly where ASR is least informative: among attacks that succeed for genuinely different reasons. Across four aligned LLMs and three jailbreak paradigms, attacks with nearly identical ASR land at clearly different points on the plane: the same model can fail through different temporal patterns. The geometry matches refusal-direction probes from hidden states on most conditions, with one model showing the limit of our fixed-lexicon approach. A simple early-stop rule derived from TLO cuts successful jailbreaks by more than half, without false alarms on plain benign queries. Safety evaluation should report when and how a failure unfolds, not only whether it occurred. TLO makes the first two observable from logits alone.

As LLMs gain stronger reasoning capabilities, their extended chain-of-thought introduces new degrees of complexity for defending against adversarial jailbreaks and prompt injection. We study consistency training, a family of fine-tuning objectives that enforce identical behavior on clean prompts and adversarial rewrites, and evaluate its two main variants, output-level (BCT) and activation-level (ACT), across five reasoning models. We formulate both methods as a prompt injection defense and find ACT to be competitive with other training-based defenses while requiring only self-supervised pairs of clean and wrapped prompts. Our experiments also generalize both techniques within the jailbreak setting, demonstrating that ACT remains more robust to adaptive attacks. We also provide mechanistic evidence that ACT's defense against jailbreaks is encoded as a roughly linear shift in activation space at the assistant-turn boundary. After ACT training, we can recover a single steering direction that controls refusal on reasoning models with minimal effect on benign inputs. We find that ACT remains robust even when the model's chain-of-thought is replaced with a compliant trace from the undefended base model, pivoting to refuse prefilled jailbreaks. Together, these results suggest that supervising internal representations is a surprisingly effective and interpretable approach to various forms of safety training in reasoning models.

Think-with-image reasoning is emerging as a new inference paradigm for large vision-language models, but its safety implications remain poorly understood. Existing systems already span multiple process designs, including direct response generation, text-only prior turn, visual-state manipulation, and explicit external image-tool invocation. In this paper, we ask which of these evaluated paradigms improves multimodal jailbreak robustness, and why. Across multiple vision-language models, explicit image-tool interaction yields the lowest attack success rates in our experiments, reducing jailbreak success by around 30% relative on average across the evaluated models. This finding is initially surprising: ASR remains low even when the returned image-tool output is manually overridden or itself unsafe-looking, but returns near direct-answering levels under text-only prior turn controls. These results indicate that the lower ASR is not explained by benign returned-image semantics or by the textual image-tool trace alone. To explain the pattern, we introduce an image-tool safety vector framework that models image-tool invocation as a residual shift in hidden representations toward a safety-relevant direction. Representation-level analyses and activation interventions support this account. Overall, our results suggest that explicit image-tool interaction is a promising design pattern for improving jailbreak robustness, while also motivating pipeline-specific safety evaluation.

Jailbreak attacks on multimodal AI systems remain underexplored, even though unsafe image generation can have more severe consequences than unsafe text and current defenses are relatively immature. We introduce PAST2HARM, a simple yet effective adaptive jailbreak framework that bypasses refusal training in state of the art multimodal text to image models. Building on prior findings that past tense reformulations can evade safeguards, PAST2HARM systematically exploits this vulnerability in multimodal generative AI. We characterize the attack along two dimensions. First, breadth: through temporal deepening, the framework incrementally strengthens historical anchoring and archival cues, eroding refusal boundaries across models with varying alignment strength. Second, depth: via iterative escalation after initial compliance, we probe the upper bound of harmful generation, measuring severity using a scalar severity jailbreak metric evaluated by a language model acting as a judge. We find that mid conversation turns form peak vulnerability windows, where harmfulness increases before plateauing and eventually undergoing semantic inversion. We evaluate PAST2HARM on three models Gemini Nano Banana Pro, GPT Image 2, and SD XL achieving attack success rates of 83 percent, 67 percent, and 100 percent in a black box, gradient free setting. Adversarial prompts also transfer across models, with cross model success rates above 50 percent. The attack elicits diverse harmful outputs, including explicit sexual content, political disinformation, historical denial narratives, hate speech, and self harm glorification. We further release a curated benchmark of prompts, reformulations, and outputs as a resource for red teaming and alignment. Our results expose fundamental brittleness in current safeguards and highlight the need for stronger multimodal safety training.

In this work, we propose BAIT (Boundary-Aware Iterative Trap), a three-step jailbreak framework that elicits malicious information through internal disclosure by target large language models (LLMs), instead of external feedback from judge LLMs. BAIT first asks the model to identify the protection boundary, then requires it to refine that boundary, and finally requests a detailed example. By expanding each step upon the model's previous responses, BAIT turns the model's own reasoning and consistency tendency into a disclosure pathway. Experiments on AdvBench, JailbreakBench, AIR-Bench, and SORRY-Bench demonstrate that BAIT consistently achieves strong attack success rates across top-tier large language models, significantly advancing conventional jailbreak baselines. Further analysis reveals that: 1) prevention-oriented framing significantly outperforms direct knowledge requests; 2) the refinement step plays a critical role in disclosure escalation; and 3) the first two steps have a certain chance of eliciting harmful content while triggering little filtering. These findings underscore the necessity of safeguarding against internally driven reasoning trajectories that progressively cross safety boundaries.

Recent defenses for safeguarding open-weight large language models (LLMs) are intended to prevent adversarial usage. Underlying these defenses is an assumption that new harmful behavior is learned through fine-tuning rather than elicited by jailbreaking the model. Yet, pretrained LLMs already encode substantial harmful knowledge across many domains, which raises an important question: can an adversary jailbreak safeguarded models, to achieve harmful usage without fine-tuning at all? In this paper, we show that open-weight safeguards are susceptible to simpler strategies that, despite being well known, have not been systematically evaluated against these safeguards. Specifically, we evaluate two low-cost attacks--abliteration and prefilling--that do not rely on gradient-based optimization. Across three harmfulness evaluation benchmarks (BeaverTails, HarmBench, and AdvBench), these attacks increase attack success rates against safeguarded open-weight models from below 10% to a range of 16%-96%. To mitigate this vulnerability, we introduce abliteration-resistant tuning (ART), which incorporates an abliteration-based objective into training. ART can be layered onto existing defenses and reduces the success rates of abliteration, prefilling, and their combination by 10%-20%. These findings indicate that the attack surface for open-weight models is broader than previously characterized, and that evaluations of safeguarding defenses should incorporate a more diverse set of attack strategies beyond adversarial fine-tuning.

Evaluating and mitigating a generative system's susceptibility to jailbreak attacks is critical to its safe deployment. Given the number of deployable systems, full per-configuration evaluation and optimization is impractical. In this paper, we formalize the behavioral geometry of a population of models that, by leveraging previously evaluated and defended models, supports both efficient susceptibility prediction and effective defense transfer across a population. We apply the framework to 79 models spanning 24 providers and to 100 system configurations of a single base model. Simple methods that use the behavioral geometry reach an AUPRC of $0.94$ for susceptibility detection with $\approx98\%$ fewer probes relative to a full evaluation. Using the behavioral geometry to select which model to transfer an optimized defense from outperforms same-provider assignment ($+2\%$, $p = 0.03$) at no additional probe cost, with a set of three models sufficient to cover the population. Results are robust to hyperparameter selection and judge.

Direct Preference Optimisation (DPO) is widely used for safety alignment in large language models. However, prior work shows it is brittle and exhibits poor out-of-distribution (OOD) generalisation. In this paper, we investigate whether Curriculum Learning can improve the robustness of DPO-based safety alignment. We propose Staged-Competence, a curriculum-based framework that organises preference data by difficulty, employs competence-based sampling, and progressively updates the reference model during training. Averaged across three model families, Staged-Competence reduces OOD harmful response rates by 16% and jailbreak attack success rates by 20%, while preserving general capabilities with near-zero over-refusal. We further show that Staged-Competence (1) matches baseline safety with only 75% of the training data and (2) yields better separation between safe and unsafe responses. Staged-Competence is agnostic to the policy optimisation loss and can extend to other DPO variants and alignment domains. Our code and data are available at https://github.com/Sandeep5500/curriculum-learning-for-safety.

Adversarial images pose a severe security threat to multimodal large language models through prompt injection. Existing defenses largely lack a principled understanding of the underlying mechanisms and struggle to balance efficiency and defense utility. In this work, we show that successful adversarial attacks do not rely on the entire image uniformly but instead depend on a small subset of critical image tokens. Based on this insight, we propose Gradient Token Masking (GTM), which localizes these tokens via gradient analysis and neutralizes them through masking. We find that attribution based on the first generated token's output probability fails when attacks preserve the predicted token. To overcome this, GTM utilizes the Hidden-State Gradient Norm score for generation-influence attribution under adversarial inputs. We prove that its ranking is consistent with that of the full adversarial loss gradient, providing a theoretical guarantee for accurate localization. Our method requires only a single forward-backward pass to identify and zero out a small number of high-scoring tokens, effectively disrupting the adversarial attack path. Extensive experiments on prompt injection and multimodal jailbreak attacks demonstrate that our approach reduces attack success rates (ASR) to near zero while preserving model utility with negligible computational overhead.

Large language model (LLM) safety classifiers such as Llama Guard are effective at detecting overtly harmful prompts but remain vulnerable to adversarial jailbreak attacks that disguise malicious intent through role-play scenarios, fictional framing, and indirect requests. We present Reflect-Guard, a method that augments LLM-based safety classifiers with chain-of-thought self-reflection capabilities through parameter-efficient fine-tuning. Our approach distills analytical reasoning from GPT-4o-mini into structured reflection annotations, then trains Llama-Guard-3-8B via QLoRA to generate logical self-reflections before issuing safety verdicts. Using only 1000 training examples and updating just 0.5% of model parameters (~42M), Reflect-Guard achieves substantial improvements on two challenging benchmarks. On WildGuardTest, F1 score improves from 0.770 to 0.842 (+7.2 pp), with recall on adversarial prompts increasing from 0.513 to 0.921 (+40.8 pp). On JailbreakBench, the attack success rate drops from 10.3% to 1.8%, representing an 82.5% relative reduction. These gains are especially pronounced on adversarial inputs, where the explicit reasoning step enables the model to see through obfuscation techniques that defeat standard pattern-matching approaches. Our results demonstrate that teaching safety classifiers to reason about adversarial intent, rather than simply classify surface patterns, is a promising direction for robust LLM safety.

Fine-tuning-as-a-Service (FaaS) enables personalization of large language models (LLMs), but it can weaken safety-alignment under harmful fine-tuning attacks. Recent work has shown that activating harmful-behavior modules during fine-tuning can prevent models from learning undesired behaviors, but its mechanism remains unclear. In this paper, we revisit temporary jailbreaking as a defense against harmful fine-tuning and provide a gradient-level analysis showing that it saturates safety-degrading gradients while preserving benign task-relevant gradients. Based on this insight, we propose a Buffer-and-Reinforce fine-tuning framework that buffers harmful updates during user fine-tuning and reinforces safety after adaptation. Specifically, BufferLoRA induces temporary jailbreaking as a removable adapter to reduce harmful updates during user fine-tuning. After adaptation, ReinforceLoRA, trained to recover refusal behavior under the temporarily jailbroken state, is integrated with UserLoRA via QR decomposition-based merging to reinforce safety while preserving user-task performance. Extensive experiments show that our framework achieves superior safety and utility with no additional safety data during user fine-tuning and minimal computational cost.

Jailbreak prompts can trigger harmful completions on aligned LLMs, In accordance, safety steering has been proposed: test-time activation interventions that steer jailbreak activations to trigger refusal while preserving benign utility. However, existing steering methods are fundamentally supervised and tied to a static, limited training set, whereas real jailbreaks evolve and are often out-of-distributed from the training set, leading to failures on unseen attacks. In this paper, we tackle the failure on unseen jailbreaks problem, base on unsupervised latent direction discovery. We propose a bi-level adversarial training framework for zero-shot jailbreak defense. In the inner step, we simulate diverse jail-broken activations by extrapolating from refusal-state harmful-request activations via unsupervised latent direction discovery, which expands the coverage of real jailbreak activation subspaces. In the outer step, we train a potential-induced steering field to push these adversarial jailbroken states into refusal regions while keeping benign unchanged. Across three LLMs and six classical jailbreak families, our method achieves strong defense with attack success rates mostly below 5%, and rising subspace coverage throughout training helps explain the improved generalization.

Large Reasoning Models (LRMs) have demonstrated remarkable capabilities in reasoning and generation tasks and are increasingly deployed in real-world applications. However, their explicit chain-of-thought (CoT) mechanism introduces new security risks, making them particularly vulnerable to jailbreak attacks. Existing approaches often rely on static CoT templates to elicit harmful outputs, but such fixed designs suffer from limited diversity, adaptability, and effectiveness. To overcome these limitations, we propose an adaptive evolutionary CoT jailbreak framework, called AE-CoT. Specifically, the method first rewrites harmful goals into mild prompts with teacher role-play and decomposes them into semantically coherent reasoning fragments to construct a pool of CoT jailbreak candidates. Then, within a structured representation space, we perform multi-generation evolutionary search, where candidate diversity is expanded through fragment-level crossover and a mutation strategy with an adaptive mutation-rate control mechanism. An independent scoring model provides graded harmfulness evaluations, and high-scoring candidates are further enhanced with a harmful CoT template to induce more destructive generations. Extensive experiments across multiple models and datasets demonstrate the effectiveness of the proposed AE-CoT, consistently outperforming state-of-the-art jailbreak methods.

The attack surface of a multimodal large language model (MLLM) is language-dependent in ways that reveal the mechanistic structure of alignment failures. We present the first systematic cross-lingual, multimodal red-teaming study comparing jailbreak vulnerability in US English (en-US) and Mexican Spanish (es-MX) across four frontier MLLMs: Claude Sonnet 4.5, GPT-5, Pixtral Large, and Qwen Omni. Using a fixed adversarial benchmark of 363 diverse prompt scenarios administered in text-only and multimodal conditions, we collected 52,272 harm ratings and binary attack success judgements from matched panels of nine native-speaker annotators per language group. Our central finding is that language does not scale vulnerability uniformly. Bayesian mixed-effects analyses reveal that linguistic framing attacks such as role-play become substantially less effective under Spanish prompting, while visually explicit multimodal attacks become more effective, which directly implicates the prompt-language interface rather than global annotator leniency. This dissociation indicates that linguistic and visual alignment failures operate through distinct mechanisms, and that switching language is sufficient to expose that separation. The practical consequence is that safety rankings are not preserved across languages. Qwen Omni overtakes Pixtral Large as the most vulnerable model among es-MX participants, a rank reversal no scalar correction of English-condition scores could recover, and absolute attack success rates have declined across model generations without closing the gaps between them. These findings demonstrate that safety evaluation frameworks treating language and modality as independent dimensions fundamentally misspecify the attack surface of globally deployed MLLMs, and must be redesigned accordingly.

Aligned models can misbehave in several ways: they are often sycophantic, fall victim to jailbreaks, or fail to include appropriate safety warnings. Consistency training is a promising new alignment paradigm to mitigate such failures by training invariants into the model using contrastive input pairs. Existing consistency training procedures generate the supervision signal once, offline, and use supervised fine-tuning (SFT) to update the model. Unfortunately, the resulting models tend to merely memorize the surface forms of the training distribution and thus generalize poorly and regress in their capabilities. We introduce On-Policy Consistency Training (OPCT), a new consistency training approach where the objective is computed over the model's own responses to prompts, supervised by itself conditioned on corresponding contrastive prompts. We evaluate OPCT on three safety axes: sycophancy, jailbreaking, and safety awareness. Across three model families, OPCT outperforms its SFT counterpart on all safety desiderata. It nearly halves the sycophancy rate relative to baseline (8.1% vs. 15.4%, compared to 11.2% for SFT). Under an adaptive per-target attacker, OPCT holds jailbreak defense success near 99% on held-out jailbreak behaviors, whereas SFT achieves 87% on average. On safety awareness, OPCT outperforms SFT in two out of three models, and matches it on the other. OPCT also largely avoids the capability regressions that SFT induces, such as a 28-point drop on MATH-500. Our results suggest that consistency training is best implemented as OPCT rather than as SFT, especially when generalization beyond the training distribution is desired.

As large language model (LLM)-based agents increasingly participate in online discourse, red-teaming their capacity to support political influence campaigns is critical for information integrity. In pursuit of this goal, we focus on locally deployed open-source LLMs, as opposed to frontier API-only models, given their superior alignment with the operational constraints of privacy-conscious malicious actors deployed in social media environments. We introduce an empirical red-teaming framework for measuring LLM Overton Windows (OWs), defined as the range of political opinions a model can reliably express on controversial topics, and for quantifying how simple natural-language jailbreaks expand that range. We evaluate more than 30 LLMs spanning 10 model families and five countries of origin. We find systematic asymmetries in political expressivity: open-source LLMs are typically more willing to generate left-leaning social media content, OWs tend to contract inversely to model size, and regional differences are substantial despite uneven representation in the open-source ecosystem. Jailbreak potency also varies sharply across model families, motivating a workflow for identifying effective combinations of jailbreak techniques. Taken together, our results establish a practical framework for auditing the political steerability of open-source LLMs and for helping future researchers design stronger countermeasures against LLM-enabled influence campaigns.

Jailbreak attacks expose a persistent gap between the intended safety behavior of aligned large language models and their behavior under adversarial prompting. Existing automated methods are increasingly effective but each commits to a single attack family (e.g., one refinement loop, one tree search, one mutation space, or one strategy library) and no single family dominates: the best-performing method shifts across target models and harm categories, suggesting complementary strengths that per-prompt composition could exploit. We introduce LASH (LLM Adaptive Semantic Hybridization), a black-box framework that treats outputs from multiple base attacks as reusable seed prompts and adaptively composes them for each target request. Given a seed pool, LASH searches over seed subsets and softmax-normalized mixture weights; a composition module synthesizes a single candidate prompt, and a derivative-free genetic optimizer updates the weights using black-box target feedback and a two-stage fitness function combining keyword-based refusal detection with LLM-judge scoring. On JailbreakBench, which contains 100 harmful prompts across 10 categories, we evaluate LASH on six common target models. LASH achieves an average attack success rate of 84.5% under keyword-based evaluation and 74.5% under two-stage evaluation, where responses are first filtered for refusals and then scored by an LLM judge for whether they substantively fulfill the original harmful request. LASH outperforms five state-of-the-art baselines on both metrics with only 30 mean target queries. LASH also remains competitive under three defense mechanisms and induces more success-like internal representations. These results suggest that adaptive composition across heterogeneous jailbreak strategies is a promising direction for black-box red-teaming.

While Large Language Models (LLMs) demonstrate remarkable capabilities, they remain susceptible to sophisticated, multi-step jailbreak attacks that circumvent conventional surface-level safety alignment by exploiting the internal generation process. To address these vulnerabilities, we propose Reflector, a principled two-stage framework that internalizes self-reflection within the generation trajectory. Reflector first leverages teacher-guided generation to produce high-quality reflection data for supervised fine-tuning (SFT), establishing structured reflection patterns. It subsequently uses Reinforcement Learning (RL) with outcome-driven and reward-validity supervision to instill robust, autonomous self-reflection capabilities. Empirical results show that Reflector achieves Defense Success Rates (DSR) exceeding 90% against complex indirect attacks while generalizing robustly across diverse threat scenarios. Notably, the framework enhances both task-specific and general utility, yielding a 5.85% gain on GSM8K alongside improved performance on knowledge-intensive benchmarks. By internalizing trajectory-level safety, Reflector overcomes the fundamental limitations of surface alignment without significant computational overhead, offering an efficient and scalable solution for the development of safe and capable LLMs.

Large Reasoning Models (LRMs) have demonstrated remarkable capabilities in solving complex problems by generating structured, step-by-step reasoning content. However, exposing a model's internal reasoning process introduces additional safety risks; for example, recent studies show that LRMs are more vulnerable to jailbreak attacks than standard LLMs. In this paper, we investigate jailbreak attacks on LRMs and reveal that the attack success rate (ASR) is closely correlated with LRMs' attention patterns. Specifically, successful jailbreaks tend to assign lower attention to harmful tokens in the input prompt, while allocating higher attention to those tokens in the reasoning content. Motivated by this finding, we propose a novel jailbreak method for LRMs that leverages reinforcement learning (RL) to enhance attack effectiveness, explicitly incorporating attention signals into the reward function design. In addition, we introduce diverse persuasion strategies to enrich the RL action space, which consistently improves the ASR. Extensive experiments on five open-source and closed-source LRMs across three benchmarks demonstrate that our method achieves substantially higher ASR, outperforming existing approaches in terms of effectiveness, efficiency, and transferability.

Recent work has demonstrated the potential of contrastive steering for jailbreaking Large Language Models (LLMs). However, existing methods rely on limited and inherently biased contrastive prompts and require laborious manual tuning of steering strength, limiting their robustness and effectiveness. In this paper, we leverage the idea of model extraction to guide the learned steering vectors to approximate the ideal one and propose tuning the steering strength adaptively based on contrastive activations' statistics. Experiments demonstrate that our method notably improves the effectiveness and robustness of probe-based steering, without any extra contrastive prompts or laborious manual tuning. Being an attack paper, this paper focuses on revealing the breakdown of fortified LLMs, raising the average harmfulness score from 6% to 70%. Our code is available at https://github.com/fhdnskfbeuv/adaptiveSteering.

As jailbreaks, adversarially crafted inputs that bypass safety constraints, continue to be discovered in Large Language Models, practitioners increasingly rely on fine-tuning as a defensive strategy. Yet the theoretical foundations underlying this robustness fine-tuning remain underexplored. We introduce a game-theoretic framework in which the interaction between an evaluator (auditing the model for jailbreaks) and a trainer is formalized as a two-player game. A key feature of our approach is the use of group actions, a mathematical structure that captures symmetries and transformations, to formally represent data augmentation. The simplest non-trivial instance is the circle with cyclic translation groups, where we exhibit various regimes depending on the trainer's generalization range. Below a critical threshold, the evaluator maintains a constant miss ratio for linearly many rounds, whereas other settings can yield very different behaviors. We further provide empirical evidence supporting locality-dependence of the model: for the three model families we tested (Llama, Qwen and Mistral), we have significant evidence that fine-tuning on adversarial prompts induces only local generalization, with refusal rates on test examples highly correlated with the distance to the fine-tuning prompts. Our framework recasts the central object of adversarial evaluation: a benchmark is not a static set of prompts but an orbit under the evaluator's group action, and audit protocols that ignore trainer-side adaptation cannot distinguish a genuine fix from a memorized patch.

Large Language Models (LLMs) remain vulnerable to jailbreak attempts that circumvent safety guardrails. We investigate whether multi-turn conversations using low-resource African languages (Afrikaans, Kiswahili, isiXhosa, and isiZulu) can bypass safety mechanisms across commercial LLMs. We translated prompts from existing datasets and evaluated ChatGPT, Claude, DeepSeek, Gemini, and Grok through automated testing and human red-teaming with native speakers. Single-turn translation attacks proved ineffective, while multi-turn conversations achieved English harmful response rates from 52.7% (Claude 3.5 Haiku) to 83.6% (GPT-4o-mini), Afrikaans from 60.0% (Claude 3.5 Haiku) to 78.2% (GPT-4o-mini), and Kiswahili from 41.8% (Claude 3.5 Haiku) to 70.9% (DeepSeek). Human red-teaming increased jailbreak rates compared to automated methods. Over all evaluated languages, the average jailbreak rate increased from 59.8% to 75.8%, with improvements of +20.0% (Afrikaans), +12.7% (isiZulu), +12.3% (isiXhosa), and +1% (Kiswahili), demonstrating that poor translation quality limits jailbreak success. These findings suggest that vulnerabilities in LLMs persist in multilingual contexts and that translation quality is the critical factor determining jailbreak success in low-resource languages.

The integration of audio modality into Large Audio Language Models (LALMs) significantly expands their attack surface. Existing jailbreak paradigms predominantly treat audio as a carrier for malicious payloads, relying on semantic optimization, acoustic parameter control, or additive perturbation to embed harmful content into the audio signal. In this work, we challenge this necessity and propose a new paradigm in which the role of audio shifts from content injection to safety alignment interference. We reveal that LALM safety alignment can be compromised solely by specific Acoustic Latent Semantics (ALS), the underlying paralinguistic features intrinsic to the priors of audio generative models. Distinct from previous works that leverage explicit acoustic parameters to merely style malicious audio, we demonstrate that interference audio, benign in content but infused with specific ALS, can serve as a universal jailbreak trigger. Leveraging this insight, we propose the Acoustic Interference Attack (AIA), which decouples the attack payload from the audio. Specifically, AIA employs a set of universal, instruction-neutral interference audio, enabling standard malicious text queries to bypass safety alignment without instance-specific optimization. Extensive experiments on 10 LALMs across five datasets demonstrate that AIA achieves the state-of-the-art attack success rate. Furthermore, our interpretability analysis uncovers the inference path drift induced by AIA and identifies the inherent effective patterns within ALS, revealing the fundamental vulnerability of cross-modal alignment in LALMs.

Despite rigorous safety alignment, Large Language Models (LLMs) remain vulnerable to jailbreak attacks. Existing black-box methods often rely on heuristic templates or exhaustive trials, lacking mechanistic interpretability and query efficiency. In this study, we investigate an intrinsic vulnerability in the safety mechanisms of LLMs, where safety alignment relies on a small set of sparsely distributed attention heads, leaving much of the representational space weakly monitored. We formalize this phenomenon with a mathematical jailbreaking model that characterizes the delicate boundary of effective text obfuscation and analytically explains observed jailbreak behaviors. Guided by this model, we propose Babel, an efficient black-box attack framework that exploits the identified safety gap through systematic obfuscation sampling with iterative, feedback-driven distribution refinement, enabling reliable and high-success jailbreak attacks without access to model internals. Comprehensive evaluations on frontier commercial models demonstrate that Babel achieves state-of-the-art attack success rates and superior query efficiency. Specifically, compared to state-of-the-art methods, Babel increases the attack success rate on GPT-4o from 41.33% to 82.67% and on Claude-3-5-haiku from 38.33% to 78.33% within an average of 40 queries, providing a robust red-teaming methodology for LLMs safety research.

Multimodal Large Language Models (MLLMs) are vulnerable to jailbreak attacks, which can elicit harmful responses from MLLMs. Many MLLMs support multi-image inputs, inadvertently introducing new vulnerabilities due to less efforts on multi-image safety alignment. Previous MLLM jailbreak methods only uses a single image, which restricts the attack space: they cannot distribute harmful requests across multiple images, carry abundant information, or exploit additional visual reasoning tasks to distract MLLMs. To address these limitations, in this paper, we propose a compositional jailbreak framework, \textbf{DMN}, which leverages \textbf{D}istributed instruction, \textbf{M}ultimodal evidence and a \textbf{N}umber chain task to fully enhance the jailbreak performance. Extensive experiments show that DMN is highly effective for MLLM jailbreaking, e.g. achieving attack success rates of over 90% on GPT-4o, Gemini-2.5-pro and Claude Sonnet 4, surpassing other baselines by a large margin. This compositional, multi-image jailbreak strategy reveals fundamental weaknesses in their safety mechanisms.

Large language models exhibit safety degradation in non-English languages. Standard evaluation relies on Jailbreak Success Rate (JSR), which confounds several safety-driving factors into one, obscuring the specific cause(s) of safety failure. We introduce a latent variable model, a Multi-Group Item Response Theory (IRT) framework, that decouples safety-driving factors such as language-agnostic safety robustness ($θ$), intrinsic prompt hardness ($β$), global language processing difficulty ($γ$), and a prompt-specific cross-lingual safety gap ($τ$). Using the MultiJail dataset, we evaluate the safety robustness of 61 model configurations across 5 closed-model families and 10 languages of varying resource, aggregating a dataset of 1.9 million rows. Exploratory Factor Analysis shows safety is primarily unidimensional: models refuse different harm types mainly through a shared mechanism. Contrary to the expected trend that safety degrades largely in low-resource languages, 22 model configurations are more vulnerable in English than in low-resource languages. Low-resource languages produce more uncertain responses (high entropy) than high-resource languages. Also, high-$τ$ prompts cluster in physical harm categories like Theft and Weapons and lower-resource languages, trends validated through cross-dataset generalization. While global translation quality shows low correlation with $τ$, severe mistranslations drive high-bias outliers, as validated by native speakers. Cultural and conceptual grounding mismatches also contribute to $τ$. In predictive validation, the IRT framework achieves $\mathrm{AUC} = 0.940$, outperforming simpler baselines in predicting safe refusal of unsafe prompts. Our framework reveals concept-language vulnerabilities that aggregate metrics obscure, enabling fairer cross-lingual safety evaluation and targeted improvements in dataset construction.

Jailbreak attacks on large models have drawn growing attention due to their close ties to societal safety. This work identifies a practical yet unexplored jailbreak scenario, the wide-net-casting scenario, where an adversary can query a group of large models instead of a single one to elicit harmful outputs. Our analysis reveals substantial yet previously overlooked safety risks under this scenario. As a key part of our analysis, we further develop a novel jailbreak method tailored to the wide-net-casting scenario. With this tailored method, the jailbreak success rate can even reach 100% in some experiments when targeting the large models without additional safeguards, exposing wide-net-casting as a distinct, high-risk scenario that warrants attention in future evaluation and defense research.

Large Language Models (LLMs) and Vision Language Models (VLMs) have demonstrated impressive capabilities but remain vulnerable to jailbreaking attacks, where adversaries exploit textual or visual triggers to bypass safety guardrails. Recent defenses typically rely on safety fine-tuning or external filters to reduce the model's likelihood of producing harmful content. While effective to some extent, these methods often incur significant computational overheads and suffer from the safety utility trade-off, degrading the model's performance on benign tasks. To address these challenges, we propose EVA (Editing for Versatile Alignment against Jailbreaks), a novel framework that pioneers the application of direct model editing for safety alignment. EVA reframes safety alignment as a precise knowledge correction task. Instead of retraining massive parameters, EVA identifies and surgically edits specific neurons responsible for the model's susceptibility to harmful instructions, while leaving the vast majority of the model unchanged. By localizing the updates, EVA effectively neutralizes harmful behaviors without compromising the model's general reasoning capabilities. Extensive experiments demonstrate that EVA outperforms baselines in mitigating jailbreaks across both LLMs and VLMs, offering a precise and efficient solution for post-deployment safety alignment.

"Oh-Oh, yes, I'm the great pretender. Pretending that I'm doing well. My need is such, I pretend too much..." summarizes the state in the area of jailbreak creation and evaluation. You find this method to generate adversarial attacks proposed by a reputable institution (e.g., BoN from Anthropic or Crescendo from Microsoft Research). However, this method does not deliver on the promise claimed in the paper despite having top ASR scores against industry-grade LLMs. You successfully generate the jailbreak prompts against your target (open) model. However, the generated jailbreak prompt works against the target model with a 50% consecutive success rate (5 out of 10 attempts) despite having an 80% ASR (on paper) on the latest closed-source model (with a guardrail system)! This observation leads us to think. First, Attack Success Rate (ASR), the primary metric for LLM jailbreak benchmarking, is not a stable quantity. Second, published ASR numbers are therefore systematically inflated and incomparable across papers. Therefore, we wonder "Why a successful jailbreak prompt does not perform consistently well against a target model on which the prompts have been optimized?". To answer this question, we study the impact of stochasticity not only during attack evaluation but also during attack generation. Our evaluation includes several jailbreak attacks, models (different sizes and providers), and judges. In addition, we propose a new metric and two new frameworks (CAS-eval and CAS-gen). Our evaluation framework, CAS-eval, shows that an attack can have an ASR drop of up to 30 percentage points when a jailbreak prompt needs to succeed on more than one attempt. Thankfully, our attack generation framework (CAS-gen) improves previous jailbreak methods and helps them recover this loss of 30 percentage points!

Large language models (LLMs) are increasingly deployed in a wide range of applications, yet remain vulnerable to adversarial jailbreak attacks that circumvent their safety guardrails. Existing evaluation frameworks typically report binary success/failure metrics, failing to capture the temporal dynamics of how attacks succeed under persistent adversarial pressure. This preliminary work proposes a novel evaluation framework that applies survival analysis techniques to characterize LLM jailbreak vuln`erability. Our approach models the time-to-jailbreak as a survival outcome, enabling estimation of hazard functions, survival curves, and risk factors associated with successful attacks. We evaluate three LLMs against a subset of prompts from the HarmBench dataset spanning three attack categories. Our analysis reveals that models exhibit distinct vulnerability profiles: while one model demonstrates rapid degradation under iterative attacks, the two other models show consistent moderate vulnerability. Our framework provides actionable insights for model and LLM application developers and establishes survival analysis as a rigorous methodology for LLM safety evaluation.

Final-token safety probes monitor a single hidden state after prompt prefill, but jailbreak prompts can contain probe-visible unsafe evidence distributed across earlier user-token representations that is missed by this readout. We study this prefill-time failure mode using SafeSwitch-style probes trained only on clean harmful and benign prompts across three instruction-tuned LLMs. The probes achieve high recall on clean harmful prompts, but miss many jailbreaks and can produce false positives on safety-adjacent benign prompts. Subspace analyses suggest that missed jailbreaks differ from clean benign prompts along directions that are poorly captured by the probe's representational subspace, and increasing probe bottleneck width does not reliably resolve this mismatch. Token-level prefill analyses reveal that probe-visible unsafe evidence often appears earlier in the sequence but is not exposed at the final-token readout, while naive max-pooling over token positions overfires on safe prompts. A simple PCA-HMM trajectory model, trained only on the same clean split, recovers many final-token misses from user-content prefill trajectories without the catastrophic false-positive behavior of naive token pooling, motivating trajectory-aware hidden-state analyses as diagnostic complements to final-token probes

Large language models are increasingly deployed in safety-critical applications, where their ability to resist harmful instructions is essential. Although post-training aims to make models robust against many jailbreak strategies, recent evidence shows that stylistic reformulations, such as poetic transformation, can still bypass safety mechanisms with alarming effectiveness. This raises a central question: why do literary jailbreaks succeed? In this work, we investigate whether their effectiveness depends on specific poetic devices, on a failure to recognize literary formatting, or on deeper changes in how models process stylistically irregular prompts. We address this problem through an interpretability analysis of attention patterns. We perform input-level ablation studies to assess the contribution of individual and combinations of poetic devices; construct an interpretable vector representation of attention maps; cluster these representations and train linear probes to predict safety outcomes and literary format. Our results show that models distinguish poetic from prose formats with high accuracy, yet struggle to predict jailbreak success within each format. Clustering further reveals clear separation by literary format, but not by safety label. These findings indicate that jailbreak success is not caused by a failure to recognize poetic formatting; rather, poetic prompts induce distinct processing patterns that remain largely independent of harmful-content detection. Overall, literary jailbreaks appear to misalign large language models not through any single poetic device, but through accumulated stylistic irregularities that alter prompt processing and avoid lexical triggers considered during post-training. This suggests that robustness requires safety mechanisms that account for style-induced shifts in model behavior. We use Qwen3-14B as a representative open-weight case study.

Recent studies show that gradient-based universal image jailbreaks on vision-language models (VLMs) exhibit little or no cross-model transferability, casting doubt on the feasibility of transferable multimodal jailbreaks. We revisit this conclusion under a strictly untargeted threat model without enforcing a fixed prefix or response pattern. Our preliminary experiment reveals that refusal behavior concentrates at high-entropy tokens during autoregressive decoding, and non-refusal tokens already carry substantial probability mass among the top-ranked candidates before attack. Motivated by this finding, we propose Untargeted Jailbreak via Entropy Maximization(UJEM)-KL, a lightweight attack that maximizes entropy at these decision tokens to flip refusal outcomes, while stabilizing the remaining low-entropy positions to preserve output quality. Across three VLMs and two safety benchmarks, UJEM-KL achieves competitive white-box attack success rates and consistently improves transferability, while remaining effective under representative defenses. Our experimental results indicate that the limited transferability primarily stems from overly constrained optimization objectives.

This paper proposes a jailbreaking prompt detection method for large language models (LLMs) to defend against jailbreak attacks. Although recent LLMs are equipped with built-in safeguards, it remains possible to craft jailbreaking prompts that bypass them. We argue that such jailbreaking prompts are inherently fragile, and thus introduce an embedding disruption method to re-activate the safeguards within LLMs. Unlike previous defense methods that aim to serve as standalone solutions, our approach instead cooperates with the LLM's internal defense mechanisms by re-triggering them. Moreover, through extensive analysis, we gain a comprehensive understanding of the disruption effects and develop an efficient search algorithm to identify appropriate disruptions for effective jailbreak detection. Extensive experiments demonstrate that our approach effectively defends against state-of-the-art jailbreak attacks in white-box and black-box settings, and remains robust even against adaptive attacks.

This paper proposes a guaranteed defense method for large language models (LLMs) to safeguard against jailbreaking attacks. Drawing inspiration from the denoised-smoothing approach in the adversarial defense domain, we propose a novel smoothing-based defense method, termed Disrupt-and-Rectify Smoothing (DR-Smoothing). Specifically, we integrate a two-stage prompt processing scheme-first disrupting the input prompt, then rectifying it-into the conventional smoothing defense framework. This disrupt-and-rectify approach improves upon previous disrupt-only approaches by restoring out-of-distribution disrupted prompts to an in-distribution form, thereby reducing the risk of unpredictable LLM behavior. In addition, this two-stage scheme offers a distinct advantage in striking a balance between harmlessness and helpfulness in jailbreaking defense. Notably, we present a theoretical analysis for generic smoothing framework, offering a tight bound for the defense success probability and the requirements on the disruption strength. Our approach can defend against both token-level and prompt-level jailbreaking attacks, under both established and adaptive attacking scenarios. Extensive experiments demonstrate that our approach surpasses current state-of-the-art defense methods in terms of both harmlessness and helpfulness.

Red teaming is critical for uncovering vulnerabilities in Large Language Models (LLMs). While automated methods have improved scalability, existing approaches often rely on static heuristics or stochastic search, rendering them brittle against advanced safety alignment. To address this, we introduce Metis, a framework that reformulates jailbreaking as inference-time policy optimization within an adversarial Partially Observable Markov Decision Process (POMDP). Metis employs a self-evolving metacognitive loop to perform causal diagnosis of a target's defense logic and leverages structured feedback as a semantic gradient to refine its policy, offering enhanced interpretability through transparent reasoning traces. Extensive evaluations across 10 diverse models demonstrate that Metis achieves the strongest average Attack Success Rate (ASR) among compared methods at 89.2%, maintaining high efficacy on resilient frontier models (e.g., 76.0% on O1 and 78.0% on GPT-5-chat) where traditional baselines exhibit substantial performance degradation. By replacing redundant exploration with directed optimization, Metis reduces token costs by an average of 8.2x and up to 11.4x. Our analysis reveals that current defenses remain vulnerable to internally-steered, closed-loop reasoning trajectories under the tested settings, highlighting a critical need for next-generation defenses capable of reasoning about safety dynamically during inference.