Daily Papers

Multimodal large language models (MLLMs) excel in vision-language tasks but also pose significant risks of generating harmful content, particularly through jailbreak attacks. Jailbreak attacks refer to intentional manipulations that bypass safety mechanisms in models, leading to the generation of inappropriate or unsafe content. Detecting such attacks is critical to ensuring the responsible deployment of MLLMs. Existing jailbreak detection methods face three primary challenges: (1) Many rely on model hidden states or gradients, limiting their applicability to white-box models, where the internal workings of the model are accessible; (2) They involve high computational overhead from uncertainty-based analysis, which limits real-time detection, and (3) They require fully labeled harmful datasets, which are often scarce in real-world settings. To address these issues, we introduce a test-time adaptive framework called JAILDAM. Our method leverages a memory-based approach guided by policy-driven unsafe knowledge representations, eliminating the need for explicit exposure to harmful data. By dynamically updating unsafe knowledge during test-time, our framework improves generalization to unseen jailbreak strategies while maintaining efficiency. Experiments on multiple VLM jailbreak benchmarks demonstrate that JAILDAM delivers state-of-the-art performance in harmful content detection, improving both accuracy and speed.

Despite rich safety alignment strategies, large language models (LLMs) remain highly susceptible to jailbreak attacks, which compromise safety guardrails and pose serious security risks. Existing detection methods mainly detect jailbreak status relying on jailbreak templates present in the training data. However, few studies address the more realistic and challenging zero-shot jailbreak detection setting, where no jailbreak templates are available during training. This setting better reflects real-world scenarios where new attacks continually emerge and evolve. To address this challenge, we propose a layer-wise, module-wise, and token-wise amplification framework that progressively magnifies internal feature discrepancies between benign and jailbreak prompts. We uncover safety-relevant layers, identify specific modules that inherently encode zero-shot discriminative signals, and localize informative safety tokens. Building upon these insights, we introduce ALERT (Amplification-based Jailbreak Detector), an efficient and effective zero-shot jailbreak detector that introduces two independent yet complementary classifiers on amplified representations. Extensive experiments on three safety benchmarks demonstrate that ALERT achieves consistently strong zero-shot detection performance. Specifically, (i) across all datasets and attack strategies, ALERT reliably ranks among the top two methods, and (ii) it outperforms the second-best baseline by at least 10% in average Accuracy and F1-score, and sometimes by up to 40%.

Safety heads are lightweight classifiers attached to pretrained language models for flagging harmful inputs before generation. Their empirical detection performance has been studied, but their formal robustness properties remain largely unexplored. We ask when a State Space Model (SSM)-based safety head can be certified to produce the same prediction for all inputs within a bounded embedding-space perturbation. We prove that the answer turns on a single condition: the l_infty norm of the state transition matrix must satisfy A_infty<1 (the contraction condition), which enables exact interval bound propagation (IBP) certification for linear time-invariant classifiers. When the contraction condition holds, the reachable output interval has bounded steady-state width and examples can be certified as robustly classified. When it fails, the interval grows exponentially with sequence length and certification is impossible at any practical perturbation radius. We enforce contraction with a hinge penalty and show on toxic comment data that certified fraction improves from 41\% to 59\%, with a sharp empirical phase transition at A_infty=1 matching the theory. Applying a contraction-regularized S4 head to jailbreak detection on JailbreakBench, we achieve a zero-shot transfer to AdvBench (DR=0.994) and HarmBench (DR=0.988). A logistic regression on mean-pooled Mamba-130M embeddings matches or exceeds the S4 head on every detection metric, confirming that harmful intent is already linearly separable in the embedding space. The S4 safety head's contribution is not superior discrimination but the formal certification that no probe-based approach provides.

The adoption of large language models (LLMs) in many applications, from customer service chat bots and software development assistants to more capable agentic systems necessitates research into how to secure these systems. Attacks like prompt injection and jailbreaking attempt to elicit responses and actions from these models that are not compliant with the safety, privacy, or content policies of organizations using the model in their application. In order to counter abuse of LLMs for generating potentially harmful replies or taking undesirable actions, LLM owners must apply safeguards during training and integrate additional tools to block the LLM from generating text that abuses the model. Jailbreaking prompts play a vital role in convincing an LLM to generate potentially harmful content, making it important to identify jailbreaking attempts to block any further steps. In this work, we propose a novel approach to detect jailbreak prompts based on pairing text embeddings well-suited for retrieval with traditional machine learning classification algorithms. Our approach outperforms all publicly available methods from open source LLM security applications.

Large Language Models (LLMs) guardrail systems are designed to protect against prompt injection and jailbreak attacks. However, they remain vulnerable to evasion techniques. We demonstrate two approaches for bypassing LLM prompt injection and jailbreak detection systems via traditional character injection methods and algorithmic Adversarial Machine Learning (AML) evasion techniques. Through testing against six prominent protection systems, including Microsoft's Azure Prompt Shield and Meta's Prompt Guard, we show that both methods can be used to evade detection while maintaining adversarial utility achieving in some instances up to 100% evasion success. Furthermore, we demonstrate that adversaries can enhance Attack Success Rates (ASR) against black-box targets by leveraging word importance ranking computed by offline white-box models. Our findings reveal vulnerabilities within current LLM protection mechanisms and highlight the need for more robust guardrail systems.

Safety mechanisms for large language models (LLMs) remain predominantly English-centric, creating systematic vulnerabilities in multilingual deployment. Prior work shows that translating malicious prompts into other languages can substantially increase jailbreak success rates, exposing a structural cross-lingual security gap. We investigate whether such attacks can be mitigated through language-agnostic semantic similarity without retraining or language-specific adaptation. Our approach compares multilingual query embeddings against a fixed English codebook of jailbreak prompts, operating as a training-free external guardrail for black-box LLMs. We conduct a systematic evaluation across four languages, two translation pipelines, four safety benchmarks, three embedding models, and three target LLMs (Qwen, Llama, GPT-3.5). Our results reveal two distinct regimes of cross-lingual transfer. On curated benchmarks containing canonical jailbreak templates, semantic similarity generalizes reliably across languages, achieving near-perfect separability (AUC up to 0.99) and substantial reductions in absolute attack success rates under strict low-false-positive constraints. However, under distribution shift - on behaviorally diverse and heterogeneous unsafe benchmarks - separability degrades markedly (AUC approx 0.60-0.70), and recall in the security-critical low-FPR regime drops across all embedding models.

Jailbreaking poses a significant risk to the deployment of Large Language Models (LLMs) and Vision Language Models (VLMs). VLMs are particularly vulnerable because they process both text and images, creating broader attack surfaces. However, available resources for jailbreak detection are scarce, particularly in finance. To address this gap, we present FENCE, a bilingual (Korean-English) multimodal dataset for training and evaluating jailbreak detectors in financial applications. FENCE emphasizes domain realism through finance-relevant queries paired with image-grounded threats. Experiments with commercial and open-source VLMs reveal consistent vulnerabilities, with GPT-4o showing measurable attack success rates and open-source models displaying greater exposure. A baseline detector trained on FENCE achieves 99 percent in-distribution accuracy and maintains strong performance on external benchmarks, underscoring the dataset's robustness for training reliable detection models. FENCE provides a focused resource for advancing multimodal jailbreak detection in finance and for supporting safer, more reliable AI systems in sensitive domains. Warning: This paper includes example data that may be offensive.

Representation Engineering analyses often characterize refusal using static directions extracted from terminal or pooled representations. We ask whether this view misses how refusal is constructed across layer-token positions. Using causal tracing, we identify a Refusal Trajectory: a sparse upstream activation pattern that often persists even when attacks such as GCG suppress terminal refusal signals. Based on this observation, we propose SALO (Sparse Activation Localization Operator), a lightweight white-box detector that operates on raw hidden-state volumes from a selected layer window. Across Qwen, Llama, and Mistral models, SALO improves jailbreak detection on several attack families under a fixed XSTest-calibrated operating point. We further analyze static RepE-style baselines, ROI sensitivity, adaptive GCG attacks, and encoded-input boundary cases, clarifying both the promise and limitations of refusal-trajectory monitoring.

Safety and security remain critical concerns in AI deployment. Despite safety training through reinforcement learning with human feedback (RLHF) [ 32], language models remain vulnerable to jailbreak attacks that bypass safety guardrails. Universal jailbreaks - prefixes that can circumvent alignment for any payload - are particularly concerning. We show empirically that jailbreak detection systems face distribution shift, with detectors trained at one point in time performing poorly against newer exploits. To study this problem, we release JailbreaksOverTime, a comprehensive dataset of timestamped real user interactions containing both benign requests and jailbreak attempts collected over 10 months. We propose a two-pronged method for defenders to detect new jailbreaks and continuously update their detectors. First, we show how to use continuous learning to detect jailbreaks and adapt rapidly to new emerging jailbreaks. While detectors trained at a single point in time eventually fail due to drift, we find that universal jailbreaks evolve slowly enough for self-training to be effective. Retraining our detection model weekly using its own labels - with no new human labels - reduces the false negative rate from 4% to 0.3% at a false positive rate of 0.1%. Second, we introduce an unsupervised active monitoring approach to identify novel jailbreaks. Rather than classifying inputs directly, we recognize jailbreaks by their behavior, specifically, their ability to trigger models to respond to known-harmful prompts. This approach has a higher false negative rate (4.1%) than supervised methods, but it successfully identified some out-of-distribution attacks that were missed by the continuous learning approach.

Prior work on jailbreak detection has established the importance of adversarial robustness for LLMs but has largely focused on the model ability to resist adversarial inputs and to output safe content, rather than the effectiveness of external supervision systems. The only public and independent benchmark of these guardrails to date evaluates a narrow set of supervisors on limited scenarios. Consequently, no comprehensive public benchmark yet verifies how well supervision systems from the market perform under realistic, diverse attacks. To address this, we introduce BELLS, a Benchmark for the Evaluation of LLM Supervision Systems. The framework is two dimensional: harm severity (benign, borderline, harmful) and adversarial sophistication (direct vs. jailbreak) and provides a rich dataset covering 3 jailbreak families and 11 harm categories. Our evaluations reveal drastic limitations of specialized supervision systems. While they recognize some known jailbreak patterns, their semantic understanding and generalization capabilities are very limited, sometimes with detection rates close to zero when asking a harmful question directly or with a new jailbreak technique such as base64 encoding. Simply asking generalist LLMs if the user question is "harmful or not" largely outperforms these supervisors from the market according to our BELLS score. But frontier LLMs still suffer from metacognitive incoherence, often responding to queries they correctly identify as harmful (up to 30 percent for Claude 3.7 and greater than 50 percent for Mistral Large). These results suggest that simple scaffolding could significantly improve misuse detection robustness, but more research is needed to assess the tradeoffs of such techniques. Our results support the "bitter lesson" of misuse detection: general capabilities of LLMs are necessary to detect a diverse array of misuses and jailbreaks.

AI agents now hold spend authority and settle payments without per-action human confirmation. The resulting loss is often not a security failure: a counterparty with the correct domain, the correct settlement address and a genuinely delivered service can charge more than it should, and no check keyed on identity will see it. We present three artefacts for measuring and reducing that loss. First, a taxonomy of agentic commerce fraud that separates five observation levels (agent reasoning, wire, settlement rail, counterparty, principal) from the request-level and history-level evidence available at each, and records which levels can observe which attacks. Second, Agentic Commerce Bench (ACB), a benchmark of twenty fraud classes generated from production aggregates, 1,647 catalogued service operations and 1,068 settlements, of which six involve a counterparty that is exactly who it claims to be. Third, gordonguard, an open-source detector stack and offline harness with which an operator can audit an agent configuration, replay hostile counterparties without an account, and run the same detectors inline. Calibrating to a stated false-positive budget on clean training traffic gives a 6.5% clean flag rate, replicated across three independent generations, and leaves eight of twenty classes no better than chance. On the four classes a reasoning layer can observe, a widely used agent security scanner run over its jailbreak-detection panel scores zero on all four, while correctly scoring 1.0 on a jailbreak supplied as a control. A measured median payment of $0.007 places a hard constraint on deployment: one human review costs 143 times the value of the payment it examines.

Jailbreak attacks designed to bypass safety mechanisms pose a serious threat by prompting LLMs to generate harmful or inappropriate content, despite alignment with ethical guidelines. Crafting universal filtering rules remains difficult due to their inherent dependence on specific contexts. To address these challenges without relying on threshold calibration or model fine-tuning, this work introduces a semantic consistency analysis between successful and unsuccessful responses, demonstrating that a negation-aware scoring approach captures meaningful patterns. Building on this insight, a novel detection framework called NegBLEURT Forest is proposed to evaluate the degree of alignment between outputs elicited by adversarial prompts and expected safe behaviors. It identifies anomalous responses using the Isolation Forest algorithm, enabling reliable jailbreak detection. Experimental results show that the proposed method consistently achieves top-tier performance, ranking first or second in accuracy across diverse models using the crafted dataset, while competing approaches exhibit notable sensitivity to model and data variations.

Despite the implementation of safety alignment strategies, large language models (LLMs) remain vulnerable to jailbreak attacks, which undermine these safety guardrails and pose significant security threats. Some defenses have been proposed to detect or mitigate jailbreaks, but they are unable to withstand the test of time due to an insufficient understanding of jailbreak mechanisms. In this work, we investigate the mechanisms behind jailbreaks based on the Linear Representation Hypothesis (LRH), which states that neural networks encode high-level concepts as subspaces in their hidden representations. We define the toxic semantics in harmful and jailbreak prompts as toxic concepts and describe the semantics in jailbreak prompts that manipulate LLMs to comply with unsafe requests as jailbreak concepts. Through concept extraction and analysis, we reveal that LLMs can recognize the toxic concepts in both harmful and jailbreak prompts. However, unlike harmful prompts, jailbreak prompts activate the jailbreak concepts and alter the LLM output from rejection to compliance. Building on our analysis, we propose a comprehensive jailbreak defense framework, JBShield, consisting of two key components: jailbreak detection JBShield-D and mitigation JBShield-M. JBShield-D identifies jailbreak prompts by determining whether the input activates both toxic and jailbreak concepts. When a jailbreak prompt is detected, JBShield-M adjusts the hidden representations of the target LLM by enhancing the toxic concept and weakening the jailbreak concept, ensuring LLMs produce safe content. Extensive experiments demonstrate the superior performance of JBShield, achieving an average detection accuracy of 0.95 and reducing the average attack success rate of various jailbreak attacks to 2% from 61% across distinct LLMs.

Large Language Models (LLMs) have shown impressive performance as general purpose agents, but their abilities remain highly dependent on prompts which are hand written with onerous trial-and-error effort. We propose a simple and nonparametric solution to this problem, Automatic Prompt Optimization (APO), which is inspired by numerical gradient descent to automatically improve prompts, assuming access to training data and an LLM API. The algorithm uses minibatches of data to form natural language ``gradients'' that criticize the current prompt. The gradients are then ``propagated'' into the prompt by editing the prompt in the opposite semantic direction of the gradient. These gradient descent steps are guided by a beam search and bandit selection procedure which significantly improves algorithmic efficiency. Preliminary results across three benchmark NLP tasks and the novel problem of LLM jailbreak detection suggest that Automatic Prompt Optimization can outperform prior prompt editing techniques and improve an initial prompt's performance by up to 31\%, by using data to rewrite vague task descriptions into more precise annotation instructions.

As large language models (LLMs) diversify across modalities, capabilities, and cost profiles, the problem of intelligent request routing -- selecting the right model for each query at inference time -- has become a critical systems challenge. We present vLLM Semantic Router, a signal-driven decision routing framework for Mixture-of-Modality (MoM) model deployments. The central innovation is composable signal orchestration: the system extracts heterogeneous signal types from each request -- from sub-millisecond heuristic features (keyword patterns, language detection, context length, role-based authorization) to neural classifiers (domain, embedding similarity, factual grounding, modality) -- and composes them through configurable Boolean decision rules into deployment-specific routing policies. Different deployment scenarios -- multi-cloud enterprise, privacy-regulated, cost-optimized, latency-sensitive -- are expressed as different signal-decision configurations over the same architecture, without code changes. Matched decisions drive semantic model routing: over a dozen of selection algorithms analyze request characteristics to find the best model cost-effectively, while per-decision plugin chains enforce privacy and safety constraints (jailbreak detection, PII filtering, hallucination detection via the three-stage HaluGate pipeline). The system provides OpenAI API support for stateful multi-turn conversations, multi-endpoint and multi-provider routing across heterogeneous backends (vLLM, OpenAI, Anthropic, Azure, Bedrock, Gemini, Vertex AI), and a pluggable authorization factory supporting multiple auth providers. Deployed in production as an Envoy external processor, the architecture demonstrates that composable signal orchestration enables a single routing framework to serve diverse deployment scenarios with differentiated cost, privacy, and safety policies.

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.

We identify Harmfulness Propagation Dynamics (HPD): for harmful prompts, the projection of the last-token hidden state onto a learned harm direction rises monotonically with transformer depth, whereas benign prompts remain flat or oscillatory. This cross-layer signature reflects harmful intent as a progressively resolved semantic property: surface form appears early, while pragmatic intent consolidates later, making the trajectory shape more informative than any single-layer snapshot. Moreover, LDA-based harm directions, learned per layer, remain stable across random splits (pairwise cosine similarity >0.97), supporting the projection sequence as a reproducible structured signal. Building on HPD, we introduce \herald{} (Harmful Encoding Recognition via Activation Layer Dynamics). This lightweight input moderator extracts a seven-dimensional feature record, slope, curvature, monotonicity, onset layer, and related statistics from the cross-layer projection sequence and classifies it with a 288-parameter MLP. stores one d-dimensional direction per layer (262\,KB for a 32-layer, d{=}4096 model), requires no gradient computation during training, and adds only 2.6{times}10^{-6} prefill FLOPs at inference. Across eight prompt-harmfulness benchmarks and four model families, achieves an average F1 of 89.3 on OLMo2-7B, surpassing all tested guard models on adversarial jailbreak detection (98.4 vs.\ 96.9 F1) and outperforming prior latent-based methods by 2.3-4.1 F1 points on every backbone. Per-instance trajectories provide machine-readable audit records that reveal when and how harmfulness emerges, offering an interpretability advantage over single-layer approaches.

Defending large language models (LLMs) against jailbreak attacks is essential for their safe and reliable deployment. Existing defenses often rely on shallow pattern matching, which struggles to generalize to novel and unseen attack strategies. To address this challenge, we propose the Cognitive-Driven Defense (CDD) framework, which targets the underlying structure of jailbreak prompts by applying meta-operations, defined as basic manipulations that conceal harmful intent.CDD emulates human cognitive reasoning through a structured reasoning chain. It begins with a global perception of the prompt and follows with a localized analysis to uncover hidden manipulations. By applying supervised fine-tuning on this structured chain, the model learns to identify and reason about known manipulation patterns. To enhance generalization to unseen threats, an entropy-guided reinforcement learning algorithm (EG-GRPO) is introduced to encourage exploration of new types and variants of meta-operations. Experiments demonstrate that CDD can achieve state-of-the-art defense performance and exhibit strong generalization to unseen jailbreak attacks.

The autonomous AI agents using large language models can create undeniable values in all span of the society but they face security threats from adversaries that warrants immediate protective solutions because trust and safety issues arise. Considering the many-shot jailbreaking and deceptive alignment as some of the main advanced attacks, that cannot be mitigated by the static guardrails used during the supervised training, points out a crucial research priority for real world robustness. The combination of static guardrails in dynamic multi-agent system fails to defend against those attacks. We intend to enhance security for LLM-based agents through the development of new evaluation frameworks which identify and counter threats for safe operational deployment. Our work uses three examination methods to detect rogue agents through a Reverse Turing Test and analyze deceptive alignment through multi-agent simulations and develops an anti-jailbreaking system by testing it with GEMINI 1.5 pro and llama-3.3-70B, deepseek r1 models using tool-mediated adversarial scenarios. The detection capabilities are strong such as 94\% accuracy for GEMINI 1.5 pro yet the system suffers persistent vulnerabilities when under long attacks as prompt length increases attack success rates (ASR) and diversity metrics become ineffective in prediction while revealing multiple complex system faults. The findings demonstrate the necessity of adopting flexible security systems based on active monitoring that can be performed by the agents themselves together with adaptable interventions by system admin as the current models can create vulnerabilities that can lead to the unreliable and vulnerable system. So, in our work, we try to address such situations and propose a comprehensive framework to counteract the security issues.

Accurately determining whether a jailbreak attempt has succeeded is a fundamental yet unresolved challenge. Existing evaluation methods rely on misaligned proxy indicators or naive holistic judgments. They frequently misinterpret model responses, leading to inconsistent and subjective assessments that misalign with human perception. To address this gap, we introduce JADES (Jailbreak Assessment via Decompositional Scoring), a universal jailbreak evaluation framework. Its key mechanism is to automatically decompose an input harmful question into a set of weighted sub-questions, score each sub-answer, and weight-aggregate the sub-scores into a final decision. JADES also incorporates an optional fact-checking module to strengthen the detection of hallucinations in jailbreak responses. We validate JADES on JailbreakQR, a newly introduced benchmark proposed in this work, consisting of 400 pairs of jailbreak prompts and responses, each meticulously annotated by humans. In a binary setting (success/failure), JADES achieves 98.5% agreement with human evaluators, outperforming strong baselines by over 9%. Re-evaluating five popular attacks on four LLMs reveals substantial overestimation (e.g., LAA's attack success rate on GPT-3.5-Turbo drops from 93% to 69%). Our results show that JADES could deliver accurate, consistent, and interpretable evaluations, providing a reliable basis for measuring future jailbreak attacks.

The misuse of large language models (LLMs) has garnered significant attention from the general public and LLM vendors. In response, efforts have been made to align LLMs with human values and intent use. However, a particular type of adversarial prompts, known as jailbreak prompt, has emerged and continuously evolved to bypass the safeguards and elicit harmful content from LLMs. In this paper, we conduct the first measurement study on jailbreak prompts in the wild, with 6,387 prompts collected from four platforms over six months. Leveraging natural language processing technologies and graph-based community detection methods, we discover unique characteristics of jailbreak prompts and their major attack strategies, such as prompt injection and privilege escalation. We also observe that jailbreak prompts increasingly shift from public platforms to private ones, posing new challenges for LLM vendors in proactive detection. To assess the potential harm caused by jailbreak prompts, we create a question set comprising 46,800 samples across 13 forbidden scenarios. Our experiments show that current LLMs and safeguards cannot adequately defend jailbreak prompts in all scenarios. Particularly, we identify two highly effective jailbreak prompts which achieve 0.99 attack success rates on ChatGPT (GPT-3.5) and GPT-4, and they have persisted online for over 100 days. Our work sheds light on the severe and evolving threat landscape of jailbreak prompts. We hope our study can facilitate the research community and LLM vendors in promoting safer and regulated LLMs.

Automated red-teaming has become a crucial approach for uncovering vulnerabilities in large language models (LLMs). However, most existing methods focus on isolated safety flaws, limiting their ability to adapt to dynamic defenses and uncover complex vulnerabilities efficiently. To address this challenge, we propose Auto-RT, a reinforcement learning framework that automatically explores and optimizes complex attack strategies to effectively uncover security vulnerabilities through malicious queries. Specifically, we introduce two key mechanisms to reduce exploration complexity and improve strategy optimization: 1) Early-terminated Exploration, which accelerate exploration by focusing on high-potential attack strategies; and 2) Progressive Reward Tracking algorithm with intermediate downgrade models, which dynamically refine the search trajectory toward successful vulnerability exploitation. Extensive experiments across diverse LLMs demonstrate that, by significantly improving exploration efficiency and automatically optimizing attack strategies, Auto-RT detects a boarder range of vulnerabilities, achieving a faster detection speed and 16.63\% higher success rates compared to existing methods.

Jailbreak attacks pose a serious threat to the safety of Large Language Models (LLMs) by crafting adversarial prompts that bypass alignment mechanisms, causing the models to produce harmful, restricted, or biased content. In this paper, we propose SafeLLM, a novel unlearning-based defense framework that unlearn the harmful knowledge from LLMs while preserving linguistic fluency and general capabilities. SafeLLM employs a three-stage pipeline: (1) dynamic unsafe output detection using a hybrid approach that integrates external classifiers with model-internal evaluations; (2) token-level harmful content tracing through feedforward network (FFN) activations to localize harmful knowledge; and (3) constrained optimization to suppress unsafe behavior without degrading overall model quality. SafeLLM achieves targeted and irreversible forgetting by identifying and neutralizing FFN substructures responsible for harmful generation pathways. Extensive experiments on prominent LLMs (Vicuna, LLaMA, and GPT-J) across multiple jailbreak benchmarks show that SafeLLM substantially reduces attack success rates while maintaining high general-purpose performance. Compared to standard defense methods such as supervised fine-tuning and direct preference optimization, SafeLLM offers stronger safety guarantees, more precise control over harmful behavior, and greater robustness to unseen attacks. Moreover, SafeLLM maintains the general performance after the harmful knowledge unlearned. These results highlight unlearning as a promising direction for scalable and effective LLM safety.

Large Language Models (LLMs) are becoming a prominent generative AI tool, where the user enters a query and the LLM generates an answer. To reduce harm and misuse, efforts have been made to align these LLMs to human values using advanced training techniques such as Reinforcement Learning from Human Feedback (RLHF). However, recent studies have highlighted the vulnerability of LLMs to adversarial jailbreak attempts aiming at subverting the embedded safety guardrails. To address this challenge, this paper defines and investigates the Refusal Loss of LLMs and then proposes a method called Gradient Cuff to detect jailbreak attempts. Gradient Cuff exploits the unique properties observed in the refusal loss landscape, including functional values and its smoothness, to design an effective two-step detection strategy. Experimental results on two aligned LLMs (LLaMA-2-7B-Chat and Vicuna-7B-V1.5) and six types of jailbreak attacks (GCG, AutoDAN, PAIR, TAP, Base64, and LRL) show that Gradient Cuff can significantly improve the LLM's rejection capability for malicious jailbreak queries, while maintaining the model's performance for benign user queries by adjusting the detection threshold.

Diffusion models have recently achieved remarkable advancements in terms of image quality and fidelity to textual prompts. Concurrently, the safety of such generative models has become an area of growing concern. This work introduces a novel type of jailbreak, which triggers T2I models to generate the image with visual text, where the image and the text, although considered to be safe in isolation, combine to form unsafe content. To systematically explore this phenomenon, we propose a dataset to evaluate the current diffusion-based text-to-image (T2I) models under such jailbreak. We benchmark nine representative T2I models, including two close-source commercial models. Experimental results reveal a concerning tendency to produce unsafe content: all tested models suffer from such type of jailbreak, with rates of unsafe generation ranging from 8\% to 74\%. In real-world scenarios, various filters such as keyword blocklists, customized prompt filters, and NSFW image filters, are commonly employed to mitigate these risks. We evaluate the effectiveness of such filters against our jailbreak and found that, while current classifiers may be effective for single modality detection, they fail to work against our jailbreak. Our work provides a foundation for further development towards more secure and reliable T2I models.

With the rise of generative large language models (LLMs) like LLaMA and ChatGPT, these models have significantly transformed daily life and work by providing advanced insights. However, as jailbreak attacks continue to circumvent built-in safety mechanisms, exploiting carefully crafted scenarios or tokens, the safety risks of LLMs have come into focus. While numerous defense strategies--such as prompt detection, modification, and model fine-tuning--have been proposed to counter these attacks, a critical question arises: do these defenses compromise the utility and usability of LLMs for legitimate users? Existing research predominantly focuses on the effectiveness of defense strategies without thoroughly examining their impact on performance, leaving a gap in understanding the trade-offs between LLM safety and performance. Our research addresses this gap by conducting a comprehensive study on the utility degradation, safety elevation, and exaggerated-safety escalation of LLMs with jailbreak defense strategies. We propose USEBench, a novel benchmark designed to evaluate these aspects, along with USEIndex, a comprehensive metric for assessing overall model performance. Through experiments on seven state-of-the-art LLMs, we found that mainstream jailbreak defenses fail to ensure both safety and performance simultaneously. Although model-finetuning performs the best overall, their effectiveness varies across LLMs. Furthermore, vertical comparisons reveal that developers commonly prioritize performance over safety when iterating or fine-tuning their LLMs.

As large language models (LLMs) become increasingly prevalent, ensuring their robustness against adversarial misuse is crucial. This paper introduces the GAP (Graph of Attacks with Pruning) framework, an advanced approach for generating stealthy jailbreak prompts to evaluate and enhance LLM safeguards. GAP addresses limitations in existing tree-based LLM jailbreak methods by implementing an interconnected graph structure that enables knowledge sharing across attack paths. Our experimental evaluation demonstrates GAP's superiority over existing techniques, achieving a 20.8% increase in attack success rates while reducing query costs by 62.7%. GAP consistently outperforms state-of-the-art methods for attacking both open and closed LLMs, with attack success rates of >96%. Additionally, we present specialized variants like GAP-Auto for automated seed generation and GAP-VLM for multimodal attacks. GAP-generated prompts prove highly effective in improving content moderation systems, increasing true positive detection rates by 108.5% and accuracy by 183.6% when used for fine-tuning. Our implementation is available at https://github.com/dsbuddy/GAP-LLM-Safety.

Large Vision-Language Models (LVLMs) demonstrate exceptional performance across multimodal tasks, yet remain vulnerable to jailbreak attacks that bypass built-in safety mechanisms to elicit restricted content generation. Existing black-box jailbreak methods primarily rely on adversarial textual prompts or image perturbations, yet these approaches are highly detectable by standard content filtering systems and exhibit low query and computational efficiency. In this work, we present Cross-modal Adversarial Multimodal Obfuscation (CAMO), a novel black-box jailbreak attack framework that decomposes malicious prompts into semantically benign visual and textual fragments. By leveraging LVLMs' cross-modal reasoning abilities, CAMO covertly reconstructs harmful instructions through multi-step reasoning, evading conventional detection mechanisms. Our approach supports adjustable reasoning complexity and requires significantly fewer queries than prior attacks, enabling both stealth and efficiency. Comprehensive evaluations conducted on leading LVLMs validate CAMO's effectiveness, showcasing robust performance and strong cross-model transferability. These results underscore significant vulnerabilities in current built-in safety mechanisms, emphasizing an urgent need for advanced, alignment-aware security and safety solutions in vision-language systems.

The aligned Large Language Models (LLMs) are powerful language understanding and decision-making tools that are created through extensive alignment with human feedback. However, these large models remain susceptible to jailbreak attacks, where adversaries manipulate prompts to elicit malicious outputs that should not be given by aligned LLMs. Investigating jailbreak prompts can lead us to delve into the limitations of LLMs and further guide us to secure them. Unfortunately, existing jailbreak techniques suffer from either (1) scalability issues, where attacks heavily rely on manual crafting of prompts, or (2) stealthiness problems, as attacks depend on token-based algorithms to generate prompts that are often semantically meaningless, making them susceptible to detection through basic perplexity testing. In light of these challenges, we intend to answer this question: Can we develop an approach that can automatically generate stealthy jailbreak prompts? In this paper, we introduce AutoDAN, a novel jailbreak attack against aligned LLMs. AutoDAN can automatically generate stealthy jailbreak prompts by the carefully designed hierarchical genetic algorithm. Extensive evaluations demonstrate that AutoDAN not only automates the process while preserving semantic meaningfulness, but also demonstrates superior attack strength in cross-model transferability, and cross-sample universality compared with the baseline. Moreover, we also compare AutoDAN with perplexity-based defense methods and show that AutoDAN can bypass them effectively.

While recent advancements in large language model (LLM) alignment have enabled the effective identification of malicious objectives involving scene nesting and keyword rewriting, our study reveals that these methods remain inadequate at detecting malicious objectives expressed through context within nested harmless objectives. This study identifies a previously overlooked vulnerability, which we term Attack via Implicit Reference (AIR). AIR decomposes a malicious objective into permissible objectives and links them through implicit references within the context. This method employs multiple related harmless objectives to generate malicious content without triggering refusal responses, thereby effectively bypassing existing detection techniques.Our experiments demonstrate AIR's effectiveness across state-of-the-art LLMs, achieving an attack success rate (ASR) exceeding 90% on most models, including GPT-4o, Claude-3.5-Sonnet, and Qwen-2-72B. Notably, we observe an inverse scaling phenomenon, where larger models are more vulnerable to this attack method. These findings underscore the urgent need for defense mechanisms capable of understanding and preventing contextual attacks. Furthermore, we introduce a cross-model attack strategy that leverages less secure models to generate malicious contexts, thereby further increasing the ASR when targeting other models.Our code and jailbreak artifacts can be found at https://github.com/Lucas-TY/llm_Implicit_reference.

Small language models (SLMs) have emerged as promising alternatives to large language models (LLMs) due to their low computational demands, enhanced privacy guarantees and comparable performance in specific domains through light-weight fine-tuning. Deploying SLMs on edge devices, such as smartphones and smart vehicles, has become a growing trend. However, the security implications of SLMs have received less attention than LLMs, particularly regarding jailbreak attacks, which is recognized as one of the top threats of LLMs by the OWASP. In this paper, we conduct the first large-scale empirical study of SLMs' vulnerabilities to jailbreak attacks. Through systematically evaluation on 63 SLMs from 15 mainstream SLM families against 8 state-of-the-art jailbreak methods, we demonstrate that 47.6% of evaluated SLMs show high susceptibility to jailbreak attacks (ASR > 40%) and 38.1% of them can not even resist direct harmful query (ASR > 50%). We further analyze the reasons behind the vulnerabilities and identify four key factors: model size, model architecture, training datasets and training techniques. Moreover, we assess the effectiveness of three prompt-level defense methods and find that none of them achieve perfect performance, with detection accuracy varying across different SLMs and attack methods. Notably, we point out that the inherent security awareness play a critical role in SLM security, and models with strong security awareness could timely terminate unsafe response with little reminder. Building upon the findings, we highlight the urgent need for security-by-design approaches in SLM development and provide valuable insights for building more trustworthy SLM ecosystem.

Large Language Models (LLMs) have achieved remarkable success in various domains but remain vulnerable to adversarial jailbreak attacks. Existing prompt-defense strategies, including parameter-modifying and parameter-free approaches, face limitations in adaptability, interpretability, and customization, constraining their effectiveness against evolving threats. To address these challenges, we propose ShieldLearner, a novel paradigm that mimics human learning in defense. Through trial and error, it autonomously distills attack signatures into a Pattern Atlas and synthesizes defense heuristics into a Meta-analysis Framework, enabling systematic and interpretable threat detection. Furthermore, we introduce Adaptive Adversarial Augmentation to generate adversarial variations of successfully defended prompts, enabling continuous self-improvement without model retraining. In addition to standard benchmarks, we create a hard test set by curating adversarial prompts from the Wildjailbreak dataset, emphasizing more concealed malicious intent. Experimental results show that ShieldLearner achieves a significantly higher defense success rate than existing baselines on both conventional and hard test sets, while also operating with lower computational overhead, making it a practical and efficient solution for real-world adversarial defense.

Large Language Models are prone to off-topic misuse, where users may prompt these models to perform tasks beyond their intended scope. Current guardrails, which often rely on curated examples or custom classifiers, suffer from high false-positive rates, limited adaptability, and the impracticality of requiring real-world data that is not available in pre-production. In this paper, we introduce a flexible, data-free guardrail development methodology that addresses these challenges. By thoroughly defining the problem space qualitatively and passing this to an LLM to generate diverse prompts, we construct a synthetic dataset to benchmark and train off-topic guardrails that outperform heuristic approaches. Additionally, by framing the task as classifying whether the user prompt is relevant with respect to the system prompt, our guardrails effectively generalize to other misuse categories, including jailbreak and harmful prompts. Lastly, we further contribute to the field by open-sourcing both the synthetic dataset and the off-topic guardrail models, providing valuable resources for developing guardrails in pre-production environments and supporting future research and development in LLM safety.

Large language models (LLMs) remain susceptible to jailbreak and direct prompt-injection attacks, yet the strongest defensive filters frequently over-refuse benign queries and degrade user experience. Previous work on jailbreak and prompt injection detection such as GradSafe, detects unsafe prompts with a single "accept all" anchor token, but its threshold is brittle and it offers no deterministic guarantee that harmful content will not be emitted once decoding begins. We introduce Gradient-Controlled Decoding (GCD), a training-free guardrail that combines an acceptance anchor token ("Sure") and refusal anchor token ("Sorry") tightening the decision boundary and significantly lowering false positives. In the mitigation stage, if a prompt is flagged, GCD preset-injects one or two refusal tokens ("Sorry, I can't...") before autoregressive decoding resumes, guaranteeing first-token safety regardless of sampling strategy. On ToxicChat, XSTest-v2, and AdvBench, GCD reduces false positives by 52% vs. GradSafe at comparable recall, lowers attack success rate by up to 10% vs. the strongest decoding-only baseline, adds under 15-20 ms latency on an average on V100 instances, transfers to LLaMA-2-7B, Mixtral-8x7B, and Qwen-2-7B, and requires only 20 demonstration templates.

Guardrails are critical for the safe deployment of Large Language Models (LLMs)-powered software. Unlike traditional rule-based systems with limited, predefined input-output spaces that inherently constrain unsafe behavior, LLMs enable open-ended, intelligent interactions--opening the door to jailbreak attacks through user inputs. Guardrails serve as a protective layer, filtering unsafe prompts before they reach the LLM. However, prior research shows that jailbreak attacks can still succeed over 70% of the time, even against advanced models like GPT-4o. While guardrails such as LlamaGuard report up to 95% accuracy, our preliminary analysis shows their performance can drop sharply--to as low as 12%--when confronted with unseen attacks. This highlights a growing software engineering challenge: how to build a post-deployment guardrail that adapts dynamically to emerging threats? To address this, we propose AdaptiveGuard, an adaptive guardrail that detects novel jailbreak attacks as out-of-distribution (OOD) inputs and learns to defend against them through a continual learning framework. Through empirical evaluation, AdaptiveGuard achieves 96% OOD detection accuracy, adapts to new attacks in just two update steps, and retains over 85% F1-score on in-distribution data post-adaptation, outperforming other baselines. These results demonstrate that AdaptiveGuard is a guardrail capable of evolving in response to emerging jailbreak strategies post deployment. We release our AdaptiveGuard and studied datasets at https://github.com/awsm-research/AdaptiveGuard to support further research.

Large language models (LLMs) are shown to be vulnerable to jailbreaking attacks where adversarial prompts are designed to elicit harmful responses. While existing defenses effectively mitigate single-turn attacks by detecting and filtering unsafe inputs, they fail against multi-turn jailbreaks that exploit contextual drift over multiple interactions, gradually leading LLMs away from safe behavior. To address this challenge, we propose a safety steering framework grounded in safe control theory, ensuring invariant safety in multi-turn dialogues. Our approach models the dialogue with LLMs using state-space representations and introduces a novel neural barrier function (NBF) to detect and filter harmful queries emerging from evolving contexts proactively. Our method achieves invariant safety at each turn of dialogue by learning a safety predictor that accounts for adversarial queries, preventing potential context drift toward jailbreaks. Extensive experiments under multiple LLMs show that our NBF-based safety steering outperforms safety alignment, prompt-based steering and lightweight LLM guardrails baselines, offering stronger defenses against multi-turn jailbreaks while maintaining a better trade-off among safety, helpfulness and over-refusal. Check out the website here https://sites.google.com/view/llm-nbf/home.

Jailbreaking is an emerging adversarial attack that bypasses the safety alignment deployed in off-the-shelf large language models (LLMs) and has evolved into multiple categories: human-based, optimization-based, generation-based, and the recent indirect and multilingual jailbreaks. However, delivering a practical jailbreak defense is challenging because it needs to not only handle all the above jailbreak attacks but also incur negligible delays to user prompts, as well as be compatible with both open-source and closed-source LLMs. Inspired by how the traditional security concept of shadow stacks defends against memory overflow attacks, this paper introduces a generic LLM jailbreak defense framework called SelfDefend, which establishes a shadow LLM as a defense instance (in detection state) to concurrently protect the target LLM instance (in normal answering state) in the normal stack and collaborate with it for checkpoint-based access control. The effectiveness of SelfDefend builds upon our observation that existing LLMs can identify harmful prompts or intentions in user queries, which we empirically validate using mainstream GPT-3.5/4 models against major jailbreak attacks. To further improve the defense's robustness and minimize costs, we employ a data distillation approach to tune dedicated open-source defense models. When deployed to protect GPT-3.5/4, Claude, Llama-2-7b/13b, and Mistral, these models outperform seven state-of-the-art defenses and match the performance of GPT-4-based SelfDefend, with significantly lower extra delays. Further experiments show that the tuned models are robust to adaptive jailbreaks and prompt injections.

Embodied Artificial Intelligence (Embodied AI) integrates perception, cognition, planning, and interaction into agents that operate in open-world, safety-critical environments. As these systems gain autonomy and enter domains such as transportation, healthcare, and industrial or assistive robotics, ensuring their safety becomes both technically challenging and socially indispensable. Unlike digital AI systems, embodied agents must act under uncertain sensing, incomplete knowledge, and dynamic human-robot interactions, where failures can directly lead to physical harm. This survey provides a comprehensive and structured review of safety research in embodied AI, examining attacks and defenses across the full embodied pipeline, from perception and cognition to planning, action and interaction, and agentic system. We introduce a multi-level taxonomy that unifies fragmented lines of work and connects embodied-specific safety findings with broader advances in vision, language, and multimodal foundation models. Our review synthesizes insights from over 400 papers spanning adversarial, backdoor, jailbreak, and hardware-level attacks; attack detection, safe training and robust inference; and risk-aware human-agent interaction. This analysis reveals several overlooked challenges, including the fragility of multimodal perception fusion, the instability of planning under jailbreak attacks, and the trustworthiness of human-agent interaction in open-ended scenarios. By organizing the field into a coherent framework and identifying critical research gaps, this survey provides a roadmap for building embodied agents that are not only capable and autonomous but also safe, robust, and reliable in real-world deployment.

Over the past year, the vLLM Semantic Router project has released a series of work spanning: (1) core routing mechanisms -- signal-driven routing, context-length pool routing, router performance engineering, policy conflict detection, low-latency embedding models, category-aware semantic caching, user-feedback-driven routing adaptation, hallucination detection, and hierarchical content-safety classification for privacy and jailbreak protection; (2) fleet optimization -- fleet provisioning and energy-efficiency analysis; (3) agentic and multimodal routing -- multimodal agent routing, tool selection, CUA security, and multi-turn context memory and safety; (4) governance and standards -- inference routing protocols and multi-provider API extensions. Each paper tackled a specific problem in LLM inference, but the problems are not independent; for example, fleet provisioning depends on the routing policy, which depends on the workload mix, shifting as organizations adopt agentic and multimodal workloads. This paper distills those results into the Workload-Router-Pool (WRP) architecture, a three-dimensional framework for LLM inference optimization. Workload characterizes what the fleet serves (chat vs. agent, single-turn vs. multi-turn, warm vs. cold, prefill-heavy vs. decode-heavy). Router determines how each request is dispatched (static semantic rules, online bandit adaptation, RL-based model selection, quality-aware cascading). Pool defines where inference runs (homogeneous vs. heterogeneous GPU, disaggregated prefill/decode, KV-cache topology). We map our prior work onto a 3x3 WRP interaction matrix, identify which cells we have covered and which remain open, and propose twenty-one concrete research directions at the intersections, each grounded in our prior measurements, tiered by maturity from engineering-ready to open research.

Ensuring safe, policy-compliant outputs from large language models requires real-time content moderation that can scale across multiple safety dimensions. However, state-of-the-art guardrail models rely on autoregressive decoders with 7B--27B parameters, reformulating what is fundamentally a classification problem as sequential text generation, a design choice that incurs high latency and scales poorly to multi-aspect evaluation. In this work, we introduce GLiGuard, a 0.3B-parameter schema-conditioned bidirectional encoder adapted from GLiNER2 for LLM content moderation. The key idea is to encode task definitions and label semantics directly into the input sequence as structured token schemas, enabling simultaneous evaluation of prompt safety, response safety, refusal detection, 14 fine-grained harm categories, and 11 jailbreak strategies in a single non-autoregressive forward pass. This schema-conditioned design lets supported task and label blocks be composed directly in the input schema at inference time. Across nine established safety benchmarks, GLiGuard achieves F1 scores competitive with 7B--27B decoder-based guards despite being 23--90times smaller, while delivering up to 16times higher throughput and 17times lower latency. These results suggest that compact bidirectional encoders can approach the accuracy of much larger guard models while drastically reducing inference cost. Code and models are available at https://github.com/fastino-ai/GLiGuard.

Large Language Models (LLMs) face threats from jailbreak prompts. Existing methods for detecting jailbreak prompts are primarily online moderation APIs or finetuned LLMs. These strategies, however, often require extensive and resource-intensive data collection and training processes. In this study, we propose GradSafe, which effectively detects jailbreak prompts by scrutinizing the gradients of safety-critical parameters in LLMs. Our method is grounded in a pivotal observation: the gradients of an LLM's loss for jailbreak prompts paired with compliance response exhibit similar patterns on certain safety-critical parameters. In contrast, safe prompts lead to different gradient patterns. Building on this observation, GradSafe analyzes the gradients from prompts (paired with compliance responses) to accurately detect jailbreak prompts. We show that GradSafe, applied to Llama-2 without further training, outperforms Llama Guard, despite its extensive finetuning with a large dataset, in detecting jailbreak prompts. This superior performance is consistent across both zero-shot and adaptation scenarios, as evidenced by our evaluations on ToxicChat and XSTest. The source code is available at https://github.com/xyq7/GradSafe.

The integration of additional modalities increases the susceptibility of large vision-language models (LVLMs) to safety risks, such as jailbreak attacks, compared to their language-only counterparts. While existing research primarily focuses on post-hoc alignment techniques, the underlying safety mechanisms within LVLMs remain largely unexplored. In this work , we investigate whether LVLMs inherently encode safety-relevant signals within their internal activations during inference. Our findings reveal that LVLMs exhibit distinct activation patterns when processing unsafe prompts, which can be leveraged to detect and mitigate adversarial inputs without requiring extensive fine-tuning. Building on this insight, we introduce HiddenDetect, a novel tuning-free framework that harnesses internal model activations to enhance safety. Experimental results show that {HiddenDetect} surpasses state-of-the-art methods in detecting jailbreak attacks against LVLMs. By utilizing intrinsic safety-aware patterns, our method provides an efficient and scalable solution for strengthening LVLM robustness against multimodal threats. Our code will be released publicly at https://github.com/leigest519/HiddenDetect.

Large Language Models (LLMs) have demonstrated remarkable generative capabilities. However, their susceptibility to misuse has raised significant safety concerns. While post-training safety alignment methods have been widely adopted, LLMs remain vulnerable to malicious instructions that can bypass safety constraints. Recent efforts have introduced inference-time safety reasoning (system-2 alignment), where LLMs conduct a reasoning process to perform safety verification before final response. We show, however, that these checks are driven by ad-hoc reasoning that diverges from the structured human process, where they first discern a user's true intent, then evaluate the associated risk based on the true intent. Consequently, these defenses remain vulnerable to sophisticated jailbreak prompts that cloak harmful goals in seemingly benign language. To build secure and safe LLMs, we propose a reasoning-based safety alignment framework, ARMOR, that replaces the ad-hoc chains of thought reasoning process with human-aligned, structured one. At inference, ARMOR (1) detects likely jailbreak strategies, (2) extracts the user's core intent while discarding deceptive instructions, and (3) applies a policy-grounded safety analysis to the purified request. ARMOR is evaluated on adaptive jailbreak attacks and multiple safety benchmarks, and a test-time scaling is conducted to further improve its performance. Results demonstrate that ARMOR significantly enhances the robustness against state-of-the-art adaptive jailbreak attacks and outperforms recent reasoning-based aligned models across various safety benchmarks.

Jailbreak attacks reveal a persistent weakness in aligned Large Language Models: carefully crafted prompts can elicit policy-violating responses despite safety training. While most defenses operate at the prompt or output level, it remains unclear how harmful intent is encoded within the model's internal representations. We investigate this question by analyzing token-level predictive entropy trajectories across layers of a frozen LLM using the logit lens. We find that static aggregate statistics of prompt-level entropy (e.g., mean, variance) carry little discriminative signal, whereas features capturing how entropy evolves across token positions, such as monotonic rank-based trend scores, are substantially more informative. Importantly, this signal is not uniform across model depth: it is concentrated in intermediate layers and degrades at the final layer, indicating that jailbreak-relevant structure is most pronounced in mid-network representations rather than at the output head. Across multiple models (Llama, Qwen, Gemma) and adversarial benchmarks, these entropy dynamics provide architecture-consistent separation without additional training. Together, our findings show that jailbreak behavior is reflected in structured intermediate uncertainty dynamics, clarifying both which entropy-derived features encode harmful intent and where in the network that signal is most pronounced.

Prompt attacks, including jailbreaks and prompt injections, pose a critical security risk to Large Language Model (LLM) systems. In production, guardrails must mitigate these attacks under strict low-latency constraints, resulting in a deployment gap in which lightweight classifiers and rule-based systems struggle to generalize under distribution shift, while high-capacity LLM-based judges remain too slow or costly for live enforcement. In this work, we examine whether lightweight, general-purpose LLMs can reliably serve as security judges under real-world production constraints. Through careful prompt and output design, lightweight LLMs are guided through a structured reasoning process involving explicit intent decomposition, safety-signal verification, harm assessment, and self-reflection. We evaluate our method on a curated dataset combining benign queries from real-world chatbots with adversarial prompts generated via automated red teaming (ART), covering diverse and evolving patterns. Our results show that general-purpose LLMs, such as gemini-2.0-flash-lite-001, can serve as effective low-latency judges for live guardrails. This configuration is currently deployed in production as a centralized guardrail service for public service chatbots in Singapore. We additionally evaluate a Mixture-of-Models (MoM) setting to assess whether aggregating multiple LLM judges improves prompt-attack detection performance relative to single-model judges, with only modest gains observed.

Large language models (LLMs) are vulnerable to unsafe training data that even small amounts of unsafe data can lead to harmful model behaviors. Detecting and filtering such unsafe training data is essential for trustworthy model development. Current state-of-the-art (SOTA) approaches typically rely on training moderation classifiers which requires significant computational overhead and are limited to predefined taxonomies, making them less adaptable to evolving safety concerns. Moreover, these classifiers lack insight into the training process, limiting their effectiveness in filtering unsafe data. To address these limitations, we propose DABUF, leveraging data attribution to detect and filter unsafe training data by attributing harmful model outputs to influential training data points. DABUF enables flexible identification of various unsafe data types without predefined taxonomies. However, in practice, model outputs can be complex with combined safe linguistic features and unsafe content, leading to reduced attribution accuracy. In such cases, DABUF will integrate moderation classifiers to identify a minimal subset of unsafe training data for targeted attribution (such as jailbreak). When model outputs are relatively straightforward, DABUF uses model outputs directly as the attribution targets. We evaluate the performance on two different tasks: in filtering jailbreaking training data and in identifying and mitigating gender bias. DABUF outperforms SOTA approaches by up to 7.5\% in detection AUPRC in jailbreaking scenarios, and 44.1\% in detecting gender bias. Moreover, retraining on DABUF-filtered data leads to higher model safety across experiments, underscoring its versatility in addressing a broad spectrum of unsafe data issues.

Speech Language Models (SLMs) enable natural interactions via spoken instructions, which more effectively capture user intent by detecting nuances in speech. The richer speech signal introduces new security risks compared to text-based models, as adversaries can better bypass safety mechanisms by injecting imperceptible noise to speech. We analyze adversarial attacks and find that SLMs are substantially more vulnerable to jailbreak attacks, which can achieve a perfect 100% attack success rate in some instances. To improve security, we propose post-hoc patching defenses used to intervene during inference by modifying the SLM's activations that improve robustness up to 99% with (i) negligible impact on utility and (ii) without any re-training. We conduct ablation studies to maximize the efficacy of our defenses and improve the utility/security trade-off, validated with large-scale benchmarks unique to SLMs.

Production LLMs must handle inputs that attempt to override system instructions, bypass safety policies or elicit harmful responses. A common mitigation is a separate guardrail model. Existing reports, however, provide little evidence on Russian prompt injection or Russian surface obfuscation. We present HiveTraceGuard-Pro, a 0.6B generative guardrail LoRA-tuned from Qwen3-0.6B. It is trained on Russian and English and uses one binary scoring rule (safe/unsafe) for the final target turn. Its training corpus pairs harmful examples, where a counterpart exists, with benign examples from the same domain and applies eight obfuscation transforms to both labels. In one harness, we compare HiveTraceGuard-Pro with thirty-four other guards on nineteen benchmark groups, sixteen of which are public. Its aggregate key is 0.7432, behind 0.7641 and 0.7552 for the two higher-scoring guards. Over the sixteen public groups alone, its key is 0.7153 and four of the thirty-four other suite guards score higher. In a fifteen-model comparison, HiveTraceGuard-Pro has the highest clean Russian robustness combined-F1 (0.88) and Russian prompt-injection recall (0.999). Both results use Russian sets assembled by our team, and at least 27.1% of the prompt-injection set overlaps the training corpus. Its 14.3 ms median latency is the lowest among those fifteen models in that run. Across the suite, FPR is 0.268 and FNR is 0.156. All reported response results use a legacy standalone-reply serialization rather than the natural assistant-role path of the shipped chat template. We release the merged weights on Hugging Face under Apache-2.0. The corpus, evaluation sets and evaluation code remain internal.

Jailbreak attacks pose a serious threat to Large Language Models (LLMs) by bypassing their safety mechanisms. A truly advanced jailbreak is defined not only by its effectiveness but, more critically, by its stealthiness. However, existing methods face a fundamental trade-off between semantic stealth (hiding malicious intent) and linguistic stealth (appearing natural), leaving them vulnerable to detection. To resolve this trade-off, we propose StegoAttack, a framework that leverages steganography. The core insight is to embed a harmful query within a benign, semantically coherent paragraph. This design provides semantic stealth by concealing the existence of malicious content and ensures linguistic stealth by maintaining the natural fluency of the cover paragraph. We evaluate StegoAttack on four state-of-the-art, safety-aligned LLMs, including GPT-5 and Gemini-3, and benchmark it against eight leading jailbreak methods. Our results show that StegoAttack achieves an average attack success rate (ASR) of 95.50%, outperforming existing baselines across all four models. Critically, its ASR drops by less than 27.00% under external detectors, while maintaining natural language distribution. This demonstrates that steganography effectively decouples linguistic and semantic stealth, thereby posing a fully concealed yet highly effective security threat. The code is available at https://github.com/GenggengSvan/StegoAttack

We introduce WildGuard -- an open, light-weight moderation tool for LLM safety that achieves three goals: (1) identifying malicious intent in user prompts, (2) detecting safety risks of model responses, and (3) determining model refusal rate. Together, WildGuard serves the increasing needs for automatic safety moderation and evaluation of LLM interactions, providing a one-stop tool with enhanced accuracy and broad coverage across 13 risk categories. While existing open moderation tools such as Llama-Guard2 score reasonably well in classifying straightforward model interactions, they lag far behind a prompted GPT-4, especially in identifying adversarial jailbreaks and in evaluating models' refusals, a key measure for evaluating safety behaviors in model responses. To address these challenges, we construct WildGuardMix, a large-scale and carefully balanced multi-task safety moderation dataset with 92K labeled examples that cover vanilla (direct) prompts and adversarial jailbreaks, paired with various refusal and compliance responses. WildGuardMix is a combination of WildGuardTrain, the training data of WildGuard, and WildGuardTest, a high-quality human-annotated moderation test set with 5K labeled items covering broad risk scenarios. Through extensive evaluations on WildGuardTest and ten existing public benchmarks, we show that WildGuard establishes state-of-the-art performance in open-source safety moderation across all the three tasks compared to ten strong existing open-source moderation models (e.g., up to 26.4% improvement on refusal detection). Importantly, WildGuard matches and sometimes exceeds GPT-4 performance (e.g., up to 3.9% improvement on prompt harmfulness identification). WildGuard serves as a highly effective safety moderator in an LLM interface, reducing the success rate of jailbreak attacks from 79.8% to 2.4%.

With the rapid advancement of Large Language Models (LLMs), the safety of LLMs has been a critical concern requiring precise assessment. Current benchmarks primarily concentrate on single-turn dialogues or a single jailbreak attack method to assess the safety. Additionally, these benchmarks have not taken into account the LLM's capability of identifying and handling unsafe information in detail. To address these issues, we propose a fine-grained benchmark SafeDialBench for evaluating the safety of LLMs across various jailbreak attacks in multi-turn dialogues. Specifically, we design a two-tier hierarchical safety taxonomy that considers 6 safety dimensions and generates more than 4000 multi-turn dialogues in both Chinese and English under 22 dialogue scenarios. We employ 7 jailbreak attack strategies, such as reference attack and purpose reverse, to enhance the dataset quality for dialogue generation. Notably, we construct an innovative assessment framework of LLMs, measuring capabilities in detecting, and handling unsafe information and maintaining consistency when facing jailbreak attacks. Experimental results across 17 LLMs reveal that Yi-34B-Chat and GLM4-9B-Chat demonstrate superior safety performance, while Llama3.1-8B-Instruct and o3-mini exhibit safety vulnerabilities.

Detecting prompt injection and jailbreak attacks is critical for deploying LLM-based agents safely. As agents increasingly process untrusted data from emails, documents, tool outputs, and external APIs, robust attack detection becomes essential. Yet current evaluation practices and production systems have fundamental limitations. We present a comprehensive analysis using a diverse benchmark of 18 datasets spanning harmful requests, jailbreaks, indirect prompt injections, and extraction attacks. We propose Leave-One-Dataset-Out (LODO) evaluation to measure true out-of-distribution generalization, revealing that the standard practice of train-test splits from the same dataset sources severely overestimates performance: aggregate metrics show an 8.4 percentage point AUC inflation, but per-dataset gaps range from 1% to 25% accuracy-exposing heterogeneous failure modes. To understand why classifiers fail to generalize, we analyze Sparse Auto-Encoder (SAE) feature coefficients across LODO folds, finding that 28% of top features are dataset-dependent shortcuts whose class signal depends on specific dataset compositions rather than semantic content. We systematically compare production guardrails (PromptGuard 2, LlamaGuard) and LLM-as-judge approaches on our benchmark, finding all three fail on indirect attacks targeting agents (7-37% detection) and that PromptGuard 2 and LlamaGuard cannot evaluate agentic tool injection due to architectural limitations. Finally, we show that LODO-stable SAE features provide more reliable explanations for classifier decisions by filtering dataset artifacts. We release our evaluation framework at https://github.com/maxf-zn/prompt-mining to establish LODO as the appropriate protocol for prompt attack detection research.

Alignment in large language models (LLMs) is used to enforce guidelines such as safety. Yet, alignment fails in the face of jailbreak attacks that modify inputs to induce unsafe outputs. In this paper, we present and evaluate a method to assess the robustness of LLM alignment. We observe that alignment embeds a safety classifier in the target model that is responsible for deciding between refusal and compliance. We seek to extract an approximation of this classifier, called a surrogate classifier, from the LLM. We develop an algorithm for identifying candidate classifiers from subsets of the LLM model. We evaluate the degree to which the candidate classifiers approximate the model's embedded classifier in benign (F1 score) and adversarial (using surrogates in a white-box attack) settings. Our evaluation shows that the best candidates achieve accurate agreement (an F1 score above 80%) using as little as 20% of the model architecture. Further, we find attacks mounted on the surrogate models can be transferred with high accuracy. For example, a surrogate using only 50% of the Llama 2 model achieved an attack success rate (ASR) of 70%, a substantial improvement over attacking the LLM directly, where we only observed a 22% ASR. These results show that extracting surrogate classifiers is a viable (and highly effective) means for modeling (and therein addressing) the vulnerability of aligned models to jailbreaking attacks.

Safety alignment is critical for LLM-powered systems. While recent LLM-powered guardrail approaches such as LlamaGuard achieve high detection accuracy of unsafe inputs written in English (e.g., ``How to create a bomb?''), they struggle with multilingual unsafe inputs. This limitation leaves LLM systems vulnerable to unsafe and jailbreak prompts written in low-resource languages such as those in Southeast Asia. This paper introduces SEALGuard, a multilingual guardrail designed to improve the safety alignment across diverse languages. It aims to address the multilingual safety alignment gap of existing guardrails and ensure effective filtering of unsafe and jailbreak prompts in LLM-powered systems. We adapt a general-purpose multilingual language model into a multilingual guardrail using low-rank adaptation (LoRA). We construct SEALSBench, a large-scale multilingual safety alignment dataset containing over 260,000 prompts in ten languages, including safe, unsafe, and jailbreak cases. We evaluate SEALGuard against state-of-the-art guardrails such as LlamaGuard on this benchmark. Our findings show that multilingual unsafe and jailbreak prompts substantially degrade the performance of the state-of-the-art LlamaGuard, which experiences a drop in Defense Success Rate (DSR) by 9% and 18%, respectively, compared to its performance on English-only prompts. In contrast, SEALGuard outperforms existing guardrails in detecting multilingual unsafe and jailbreak prompts, improving DSR by 48% over LlamaGuard and achieving the best DSR, precision, and F1-score. Our ablation study further reveals the contributions of adaptation strategies and model size to the overall performance of SEALGuard. We release our pre-trained model and benchmark at https://github.com/awsm-research/SEALGuard to support further research.

We introduce the Granite Guardian models, a suite of safeguards designed to provide risk detection for prompts and responses, enabling safe and responsible use in combination with any large language model (LLM). These models offer comprehensive coverage across multiple risk dimensions, including social bias, profanity, violence, sexual content, unethical behavior, jailbreaking, and hallucination-related risks such as context relevance, groundedness, and answer relevance for retrieval-augmented generation (RAG). Trained on a unique dataset combining human annotations from diverse sources and synthetic data, Granite Guardian models address risks typically overlooked by traditional risk detection models, such as jailbreaks and RAG-specific issues. With AUC scores of 0.871 and 0.854 on harmful content and RAG-hallucination-related benchmarks respectively, Granite Guardian is the most generalizable and competitive model available in the space. Released as open-source, Granite Guardian aims to promote responsible AI development across the community. https://github.com/ibm-granite/granite-guardian

Large language models (LLMs) have evolved from simple chatbots into autonomous agents capable of performing complex tasks such as editing production code, orchestrating workflows, and taking higher-stakes actions based on untrusted inputs like webpages and emails. These capabilities introduce new security risks that existing security measures, such as model fine-tuning or chatbot-focused guardrails, do not fully address. Given the higher stakes and the absence of deterministic solutions to mitigate these risks, there is a critical need for a real-time guardrail monitor to serve as a final layer of defense, and support system level, use case specific safety policy definition and enforcement. We introduce LlamaFirewall, an open-source security focused guardrail framework designed to serve as a final layer of defense against security risks associated with AI Agents. Our framework mitigates risks such as prompt injection, agent misalignment, and insecure code risks through three powerful guardrails: PromptGuard 2, a universal jailbreak detector that demonstrates clear state of the art performance; Agent Alignment Checks, a chain-of-thought auditor that inspects agent reasoning for prompt injection and goal misalignment, which, while still experimental, shows stronger efficacy at preventing indirect injections in general scenarios than previously proposed approaches; and CodeShield, an online static analysis engine that is both fast and extensible, aimed at preventing the generation of insecure or dangerous code by coding agents. Additionally, we include easy-to-use customizable scanners that make it possible for any developer who can write a regular expression or an LLM prompt to quickly update an agent's security guardrails.

Where inside a language model does refusal live, and does that place change when the architecture does? In a transformer, refusal is governed by a single direction in the residual stream, a finding that safety and interpretability tooling now depend on. State-space models (SSMs) route information through a recurrent update instead of attention, sharing no token-mixing mechanism with a transformer. Does the same safety representation survive this shift, or must it be rediscovered per architecture? It survives. A single rigid rotation, which can only reorient a space and not reshape it, aligns one model's representation space with another's, so the two genuinely share the representation. A harm probe trained on a transformer then flags an SSM's harmful inputs, and removing the aligned direction makes a model answer attacks it would otherwise refuse, while a random direction of the same size does far less. What is architecture-specific is not where the direction is steered but where it must be read. Each layer computes a fresh output that is then added into the residual stream, and harm is cleanly readable at this output, the write site, before the addition. A control that holds the intervention's strength fixed shows that what matters is where the direction is estimated, not where it is applied. Applied through a detector-triggered gate, this direction lowers jailbreak success in all four architecture families we test (SSM, transformer, recurrent, hybrid), and on the SSM it holds against an attacker that tunes its prompt against the defense. The gate only matches a trivial rule that returns a fixed refusal whenever the same detector fires, so what transfers across architectures is the direction itself, not defense strength. Safety tooling built on refusal therefore ports to a new architecture by re-estimating the direction at that architecture's write site, not by rebuilding it.

Optimization-based jailbreaks typically adopt the Toxic-Continuation setting in large vision-language models (LVLMs), following the standard next-token prediction objective. In this setting, an adversarial image is optimized to make the model predict the next token of a toxic prompt. However, we find that the Toxic-Continuation paradigm is effective at continuing already-toxic inputs, but struggles to induce safety misalignment when explicit toxic signals are absent. We propose a new paradigm: Benign-to-Toxic (B2T) jailbreak. Unlike prior work, we optimize adversarial images to induce toxic outputs from benign conditioning. Since benign conditioning contains no safety violations, the image alone must break the model's safety mechanisms. Our method outperforms prior approaches, transfers in black-box settings, and complements text-based jailbreaks. These results reveal an underexplored vulnerability in multimodal alignment and introduce a fundamentally new direction for jailbreak approaches.

A novel hack involving Large Language Models (LLMs) has emerged, exploiting adversarial suffixes to deceive models into generating perilous responses. Such jailbreaks can trick LLMs into providing intricate instructions to a malicious user for creating explosives, orchestrating a bank heist, or facilitating the creation of offensive content. By evaluating the perplexity of queries with adversarial suffixes using an open-source LLM (GPT-2), we found that they have exceedingly high perplexity values. As we explored a broad range of regular (non-adversarial) prompt varieties, we concluded that false positives are a significant challenge for plain perplexity filtering. A Light-GBM trained on perplexity and token length resolved the false positives and correctly detected most adversarial attacks in the test set.

As large language models (LLMs) become integral to various applications, ensuring both their safety and utility is paramount. Jailbreak attacks, which manipulate LLMs into generating harmful content, pose significant challenges to this balance. Existing defenses, such as prompt engineering and safety fine-tuning, often introduce computational overhead, increase inference latency, and lack runtime flexibility. Moreover, overly restrictive safety measures can degrade model utility by causing refusals of benign queries. In this paper, we introduce Jailbreak Antidote, a method that enables real-time adjustment of LLM safety preferences by manipulating a sparse subset of the model's internal states during inference. By shifting the model's hidden representations along a safety direction with varying strengths, we achieve flexible control over the safety-utility balance without additional token overhead or inference delays. Our analysis reveals that safety-related information in LLMs is sparsely distributed; adjusting approximately 5% of the internal state is as effective as modifying the entire state. Extensive experiments on nine LLMs (ranging from 2 billion to 72 billion parameters), evaluated against ten jailbreak attack methods and compared with six defense strategies, validate the effectiveness and efficiency of our approach. By directly manipulating internal states during reasoning, Jailbreak Antidote offers a lightweight, scalable solution that enhances LLM safety while preserving utility, opening new possibilities for real-time safety mechanisms in widely-deployed AI systems.

Jailbreak attacks pose persistent threats to large language models (LLMs). Current safety alignment methods have attempted to address these issues, but they experience two significant limitations: insufficient safety alignment depth and unrobust internal defense mechanisms. These limitations make them vulnerable to adversarial attacks such as prefilling and refusal direction manipulation. We introduce DeepRefusal, a robust safety alignment framework that overcomes these issues. DeepRefusal forces the model to dynamically rebuild its refusal mechanisms from jailbreak states. This is achieved by probabilistically ablating the refusal direction across layers and token depths during fine-tuning. Our method not only defends against prefilling and refusal direction attacks but also demonstrates strong resilience against other unseen jailbreak strategies. Extensive evaluations on four open-source LLM families and six representative attacks show that DeepRefusal reduces attack success rates by approximately 95%, while maintaining model capabilities with minimal performance degradation.

Safety alignment of large language models (LLMs) is mostly evaluated in English and contract-bound, leaving multilingual vulnerabilities understudied. We introduce Indic Jailbreak Robustness (IJR), a judge-free benchmark for adversarial safety across 12 Indic and South Asian languages (2.1 Billion speakers), covering 45216 prompts in JSON (contract-bound) and Free (naturalistic) tracks. IJR reveals three patterns. (1) Contracts inflate refusals but do not stop jailbreaks: in JSON, LLaMA and Sarvam exceed 0.92 JSR, and in Free all models reach 1.0 with refusals collapsing. (2) English to Indic attacks transfer strongly, with format wrappers often outperforming instruction wrappers. (3) Orthography matters: romanized or mixed inputs reduce JSR under JSON, with correlations to romanization share and tokenization (approx 0.28 to 0.32) indicating systematic effects. Human audits confirm detector reliability, and lite-to-full comparisons preserve conclusions. IJR offers a reproducible multilingual stress test revealing risks hidden by English-only, contract-focused evaluations, especially for South Asian users who frequently code-switch and romanize.

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.

The inherent risk of generating harmful and unsafe content by Large Language Models (LLMs), has highlighted the need for their safety alignment. Various techniques like supervised fine-tuning, reinforcement learning from human feedback, and red-teaming were developed for ensuring the safety alignment of LLMs. However, the robustness of these aligned LLMs is always challenged by adversarial attacks that exploit unexplored and underlying vulnerabilities of the safety alignment. In this paper, we develop a novel black-box jailbreak attack, called BitBypass, that leverages hyphen-separated bitstream camouflage for jailbreaking aligned LLMs. This represents a new direction in jailbreaking by exploiting fundamental information representation of data as continuous bits, rather than leveraging prompt engineering or adversarial manipulations. Our evaluation of five state-of-the-art LLMs, namely GPT-4o, Gemini 1.5, Claude 3.5, Llama 3.1, and Mixtral, in adversarial perspective, revealed the capabilities of BitBypass in bypassing their safety alignment and tricking them into generating harmful and unsafe content. Further, we observed that BitBypass outperforms several state-of-the-art jailbreak attacks in terms of stealthiness and attack success. Overall, these results highlights the effectiveness and efficiency of BitBypass in jailbreaking these state-of-the-art LLMs.

Conversational large language models are fine-tuned for both instruction-following and safety, resulting in models that obey benign requests but refuse harmful ones. While this refusal behavior is widespread across chat models, its underlying mechanisms remain poorly understood. In this work, we show that refusal is mediated by a one-dimensional subspace, across 13 popular open-source chat models up to 72B parameters in size. Specifically, for each model, we find a single direction such that erasing this direction from the model's residual stream activations prevents it from refusing harmful instructions, while adding this direction elicits refusal on even harmless instructions. Leveraging this insight, we propose a novel white-box jailbreak method that surgically disables refusal with minimal effect on other capabilities. Finally, we mechanistically analyze how adversarial suffixes suppress propagation of the refusal-mediating direction. Our findings underscore the brittleness of current safety fine-tuning methods. More broadly, our work showcases how an understanding of model internals can be leveraged to develop practical methods for controlling model behavior.

The vulnerability of Vision Large Language Models (VLLMs) to jailbreak attacks appears as no surprise. However, recent defense mechanisms against these attacks have reached near-saturation performance on benchmark evaluations, often with minimal effort. This dual high performance in both attack and defense raises a fundamental and perplexing paradox. To gain a deep understanding of this issue and thus further help strengthen the trustworthiness of VLLMs, this paper makes three key contributions: i) One tentative explanation for VLLMs being prone to jailbreak attacks--inclusion of vision inputs, as well as its in-depth analysis. ii) The recognition of a largely ignored problem in existing defense mechanisms--over-prudence. The problem causes these defense methods to exhibit unintended abstention, even in the presence of benign inputs, thereby undermining their reliability in faithfully defending against attacks. iii) A simple safety-aware method--LLM-Pipeline. Our method repurposes the more advanced guardrails of LLMs on the shelf, serving as an effective alternative detector prior to VLLM response. Last but not least, we find that the two representative evaluation methods for jailbreak often exhibit chance agreement. This limitation makes it potentially misleading when evaluating attack strategies or defense mechanisms. We believe the findings from this paper offer useful insights to rethink the foundational development of VLLM safety with respect to benchmark datasets, defense strategies, and evaluation methods.

Automated red teaming frameworks for Large Language Models (LLMs) have become increasingly sophisticated, yet they share a fundamental limitation: their jailbreak logic is confined to selecting, combining, or refining pre-existing attack strategies. This binds their creativity and leaves them unable to autonomously invent entirely new attack mechanisms. To overcome this gap, we introduce EvoSynth, an autonomous framework that shifts the paradigm from attack planning to the evolutionary synthesis of jailbreak methods. Instead of refining prompts, EvoSynth employs a multi-agent system to autonomously engineer, evolve, and execute novel, code-based attack algorithms. Crucially, it features a code-level self-correction loop, allowing it to iteratively rewrite its own attack logic in response to failure. Through extensive experiments, we demonstrate that EvoSynth not only establishes a new state-of-the-art by achieving an 85.5\% Attack Success Rate (ASR) against highly robust models like Claude-Sonnet-4.5, but also generates attacks that are significantly more diverse than those from existing methods. We release our framework to facilitate future research in this new direction of evolutionary synthesis of jailbreak methods. Code is available at: https://github.com/dongdongunique/EvoSynth.

Refusal refers to the functional behavior enabling safety-aligned language models to reject harmful or unethical prompts. Following the growing scientific interest in mechanistic interpretability, recent work encoded refusal behavior as a single direction in the model's latent space; e.g., computed as the difference between the centroids of harmful and harmless prompt representations. However, emerging evidence suggests that concepts in LLMs often appear to be encoded as a low-dimensional manifold embedded in the high-dimensional latent space. Motivated by these findings, we propose a novel method leveraging Self-Organizing Maps (SOMs) to extract multiple refusal directions. To this end, we first prove that SOMs generalize the prior work's difference-in-means technique. We then train SOMs on harmful prompt representations to identify multiple neurons. By subtracting the centroid of harmless representations from each neuron, we derive a set of multiple directions expressing the refusal concept. We validate our method on an extensive experimental setup, demonstrating that ablating multiple directions from models' internals outperforms not only the single-direction baseline but also specialized jailbreak algorithms, leading to an effective suppression of refusal. Finally, we conclude by analyzing the mechanistic implications of our approach.

Multimodal Large Language Models (MLLMs) have achieved impressive performance, but their safety alignment remains vulnerable to jailbreak attacks. Existing content-based jailbreaks are often inconsistent and show unsatisfying performance against the rapidly evolving MLLMs, failing to exploit non-content-based vulnerabilities. Unlike previous research, we empirically find that MLLMs exhibit a Stylistic Inconsistency between their comprehension ability and safety ability: MLLMs can robustly understand content regardless of visual style, yet their defense mechanisms can be easily bypassed by specific stylistic triggers. Based on this finding, we propose Adversarial Style Optimization (ASO), a plug-and-play enhancement module to amplify existing visual jailbreaks. ASO fine-tunes an image-editing model to superimpose an optimized stylistic modification onto a given adversarial image, using a Group Relative Policy Optimization (GRPO) agent guided by a Structurally-Tiered Reward Function that combines a logit-based signal for detecting explicit refusals with a high-fidelity semantic evaluation from a powerful judge model. Extensive experiments show that ASO significantly enhances the ASR of SOTA attacks, demonstrating that stylistic biases are a scalable vector for red-teaming MLLMs. Our code is available at https://github.com/bingjunluo/ASO.

Jailbreak attacks, which aim to cause LLMs to perform unrestricted behaviors, have become a critical and challenging direction in AI safety. Despite achieving the promising attack success rate using dictionary-based evaluation, existing jailbreak attack methods fail to output detailed contents to satisfy the harmful request, leading to poor performance on GPT-based evaluation. To this end, we propose a black-box jailbreak attack termed GeneShift, by using a genetic algorithm to optimize the scenario shifts. Firstly, we observe that the malicious queries perform optimally under different scenario shifts. Based on it, we develop a genetic algorithm to evolve and select the hybrid of scenario shifts. It guides our method to elicit detailed and actionable harmful responses while keeping the seemingly benign facade, improving stealthiness. Extensive experiments demonstrate the superiority of GeneShift. Notably, GeneShift increases the jailbreak success rate from 0% to 60% when direct prompting alone would fail.

Reasoning models generate chain-of-thought (CoT) tokens before their final output, but how this affects their vulnerability to jailbreak attacks remains unclear. While traditional language models make refusal decisions at the prompt-response boundary, we find evidence that DeepSeek-R1-Distill-Llama-8B makes these decisions within its CoT generation. We identify a linear direction in activation space during CoT token generation that predicts whether the model will refuse or comply -- termed the "caution" direction because it corresponds to cautious reasoning patterns in the generated text. Ablating this direction from model activations increases harmful compliance, effectively jailbreaking the model. We additionally show that intervening only on CoT token activations suffices to control final outputs, and that incorporating this direction into prompt-based attacks improves success rates. Our findings suggest that the chain-of-thought itself is a promising new target for adversarial manipulation in reasoning models. Code available at https://github.com/ky295/reasoning-manipulation

As AI agents become widely deployed as online services, users often rely on an agent developer's claim about how safety is enforced, which introduces a threat where safety measures are falsely advertised. To address the threat, we propose proof-of-guardrail, a system that enables developers to provide cryptographic proof that a response is generated after a specific open-source guardrail. To generate proof, the developer runs the agent and guardrail inside a Trusted Execution Environment (TEE), which produces a TEE-signed attestation of guardrail code execution verifiable by any user offline. We implement proof-of-guardrail for OpenClaw agents and evaluate latency overhead and deployment cost. Proof-of-guardrail ensures integrity of guardrail execution while keeping the developer's agent private, but we also highlight a risk of deception about safety, for example, when malicious developers actively jailbreak the guardrail. Code and demo video: https://github.com/SaharaLabsAI/Verifiable-ClawGuard

Emergent misalignment (EM) is the phenomenon where fine-tuning a language model on a narrow task leads to harmful behavior in unrelated domains. A leading mechanistic account attributes EM to persona features: latent directions acquired during pre-training that misaligned fine-tuning amplifies. We ask where these features come from: which pre-training documents activate them, and whether naturally occurring human-written text suffices to induce EM. Using Sparse Autoencoder (SAE) based model diffing across four open-weight models, we find that features related to jailbreak personas, sarcasm, deception, and manipulation are amplified by misalignment fine-tuning, while safety-relevant and assistant-identity features are suppressed. Steering individual features controls EM in both directions: it induces misalignment rates of up to 62% in aligned models -- exceeding the 35% reached by misalignment fine-tuning itself -- and re-aligns misaligned models to near-baseline misalignment rates. Attributing the causal features to a corpus of one million pre-training web documents retrieves semantically relevant narratives about villainous characters, domination, and harmful agency. However, fine-tuning on these human-written documents does not reliably induce EM, even after reformatting into assistant-style responses, whereas synthetic instruction-response pairs derived from the same content do -- and transfer across model families. Semantic relevance alone is therefore not sufficient: response structure or model-generated phrasing plays an important role in inducing EM.

Despite extensive safety-tuning, large language models (LLMs) remain vulnerable to jailbreak attacks via adversarially crafted instructions, reflecting a persistent trade-off between safety and task performance. In this work, we propose Intent-FT, a simple and lightweight fine-tuning approach that explicitly trains LLMs to infer the underlying intent of an instruction before responding. By fine-tuning on a targeted set of adversarial instructions, Intent-FT enables LLMs to generalize intent deduction to unseen attacks, thereby substantially improving their robustness. We comprehensively evaluate both parametric and non-parametric attacks across open-source and proprietary models, considering harmfulness from attacks, utility, over-refusal, and impact against white-box threats. Empirically, Intent-FT consistently mitigates all evaluated attack categories, with no single attack exceeding a 50\% success rate -- whereas existing defenses remain only partially effective. Importantly, our method preserves the model's general capabilities and reduces excessive refusals on benign instructions containing superficially harmful keywords. Furthermore, models trained with Intent-FT accurately identify hidden harmful intent in adversarial attacks, and these learned intentions can be effectively transferred to enhance vanilla model defenses. We publicly release our code at https://github.com/wj210/Intent_Jailbreak.

Large language models (LLMs) remain vulnerable to jailbreak attacks that conceal harmful intent within complex adversarial prompts. Existing defenses primarily rely on input perturbation or harmful-output suppression, but they rarely model where malicious intent resides, resulting in brittle protection and excessive over-refusal. We propose SENTINEL, a plug-and-play, generation-time jailbreak defense that reframes mitigation as an intent extraction problem. Our key insight is that instruction-tuned LLMs exhibit strong input--output semantic consistency: regardless of jailbreak complexity, generated outputs tend to align with the attacker's true intent. SENTINEL exploits this property by matching semantically aligned input--output regions to extract intention-revealing subsequences, scores these subsequences using refusal-direction projections to estimate harmfulness, and halts generation when necessary. Experiments on HarmBench across multiple LLMs show that SENTINEL reduces jailbreak success rates to close to 5\% while maintaining low over-refusal. We further demonstrate robustness to adaptive attacks and provide a mechanistic interpretation: SENTINEL re-distributes jailbreak features from alignment blind spots to aligned regions.

Safety alignment in open-weight language models is trivially removable: abliteration projects a refusal-mediating direction out of the weights in minutes, and no release-time defense we are aware of prevents it durably. What cannot be prevented can be deceived. Our defense, decoy hardening ("Fool's Gold"), concedes the refusal strip and poisons its payoff: once refusal is stripped, most answers to hazardous operational requests are confident, fluent decoys whose critical elements are falsified. Decoys are trained inside a differentiable simulation of the attack, expressing only in the attacked state; a refusal pin and benign leash hold clean-state behavior to the original. We instantiate it on seven models from five families (9B-122B, dense and mixture-of-experts). On the six models passing our pre-registered efficacy gate, 0.51-0.90 of attacked-state responses to held-out prompts are decoys, +0.27-0.84 attributable to the defense; all six stay within registered benign-behavior and capability budgets; the seventh (smaller) fails the gate (boundary case). Rates replicate on a frozen test split or untouched strata. The claim is epistemic: without independent ground truth, no observation surface we tested separates falsified answers from correct ones - on external red-team benchmarks' CBRNE-adjacent slice, the defended 122B is fatally wrong on 0.82-0.86 of matched-quality answers vs at most 0.10 undefended. Repeated sampling does not restore trust: element-wise consensus at K=64 reconstructs a fully usable procedure on 0.083-0.625 of prompts where the instrument validates, vs 0.58-0.96 undefended, with no label-free way to tell the regimes apart; on the weakest such model the claim is per-draw only. We evaluate chemical and biological hazards; the defense does not address in-context jailbreaks and protects only the initially released defended weights.

Detectors of alignment failures screen deployed language models and score alignment benchmarks. Most are generative judges that spend a decoding pass on every criterion, and classifiers that read token probabilities, such as Llama Guard, still score one fixed label per call. Jev, a model trained with reinforcement learning for calibrated decisions (RLCD), answers many typed questions about one input with calibrated probabilities in a single call. Whether it detects alignment failures has not been measured. We present RLCDAlignBench, which benchmarks Jev on ten alignment failures: sycophancy, jailbreaks, deception, prompt injection, hallucination, privacy violation, social bias, reward hacking, concealing uncertainty, and power seeking. It spans 44 benchmarks and five target models, labelled by each benchmark's scorer and, on two, by humans. Many of these failures are relational, defined against a reference, such as the user's belief or an injected instruction, that the response alone does not reveal. Our key idea is therefore to vary what Jev is asked separately from what it sees: the question's wording and answer type on one side, the fields of the input on the other. A single generic question reaches a median AUROC of 0.886 zero-shot and beats supervised baselines on most benchmarks. Question wording matters little, while context matters more, mostly through fields that encode the label. Jev matches the reference scorer's agreement with human labels, surfaces label defects in existing benchmarks, and costs 63x less than LLM-judge scorers. Code and data: https://github.com/sumleo/RLCDAlignBench.

Understanding how aligned LLMs internally represent safety is critical for diagnosing alignment vulnerabilities, as it explains why jailbreaks succeed and informs the design of robust alignment strategies. Prior work shows that aligned LLMs encode harmfulness and refusal as separable directions in the residual stream at prompt-side token positions. We show that jailbreaks succeed at prompt encoding by suppressing either the refusal or harmfulness direction before any token is generated, with distinct attack classes occupying separable regions of the harmfulness-refusal plane. Extending the analysis to response-token positions, we find that the model recognizes harmful content while it is generating that content, even when it failed to recognize the input as harmful at the prompt side. Motivated by our findings, we introduce HARC (Harmfulness-And-Refusal Coupling), a fine-tuning method that pairs the two directions across both prompt and response positions. Since the intervention is confined to the harmfulness-refusal subspace, it leaves the rest of the residual stream intact and does not degrade general capability or inflate over-refusal. Across extensive experiments, HARC achieves the strongest robustness-capability-usability trade-off among six baselines spanning the major training-time and inference-time safety methods. The harmfulness and refusal directions at prompt and response positions transfer across the five model families and two scales we tested without architecture-specific tuning.