Daily Papers

Mitigating sensitive and harmful outputs is fundamental to ensuring safe deployment of LLMs. Existing approaches typically follow two paradigms: Knowledge Deletion (KD), which erases undesirable information during training, and Distinguishable Refusal (DR), which steers models away from using sensitive knowledge during inference. Despite rapid progress, KD-based unlearning struggles with biased deletion due to suppressing specific token sequences as a substitute for complete knowledge removal, whereas DR-based unlearning risks the re-emergence of harmful knowledge because the underlying knowledge remains intact. To address these issues, we propose Distinguishable Deletion (D^2), a paradigm that restricts the response distribution in the latent representation rather than specific tokens to erase undesirable knowledge, while distinguishing it from retained knowledge, enabling a refusal mechanism to handle unlearned inputs safely and coherently. To implement D^2, we introduce an energy index that quantifies the presence of knowledge and the separation between unlearned and retained content. Mathematical and empirical analyses show that energy is both accurate and efficient, enabling Energy-based Unlearning Alignment (EUA) to enforce energy-boundary unlearning during training and apply an energy-based refusal mechanism at inference. Extensive experiments demonstrate that EUA significantly outperforms previous methods, indicating the superiority of D^2. Our code is available at https://github.com/Puning97/EUA-for-LLM-Unlearning.

Current evaluations of LLM safety predominantly rely on severity-based taxonomies to assess the harmfulness of malicious queries. We argue that this formulation requires re-examination as it assumes uniform risk across all malicious queries, neglecting Execution Likelihood--the conditional probability of a threat being realized given the model's response. In this work, we introduce Expected Harm, a metric that weights the severity of a jailbreak by its execution likelihood, modeled as a function of execution cost. Through empirical analysis of state-of-the-art models, we reveal a systematic Inverse Risk Calibration: models disproportionately exhibit stronger refusal behaviors for low-likelihood (high-cost) threats while remaining vulnerable to high-likelihood (low-cost) queries. We demonstrate that this miscalibration creates a structural vulnerability: by exploiting this property, we increase the attack success rate of existing jailbreaks by up to 2times. Finally, we trace the root cause of this failure using linear probing, which reveals that while models encode severity in their latent space to drive refusal decisions, they possess no distinguishable internal representation of execution cost, making them "blind" to this critical dimension of risk.

The ability of language models in RAG systems to selectively refuse to answer based on flawed context is critical for safety, yet remains a significant failure point. Our large-scale study reveals that even frontier models struggle in this setting, with refusal accuracy dropping below 50% on multi-document tasks, while exhibiting either dangerous overconfidence or overcaution. Static benchmarks fail to reliably evaluate this capability, as models exploit dataset-specific artifacts and memorize test instances. We introduce RefusalBench, a generative methodology that programmatically creates diagnostic test cases through controlled linguistic perturbation. Our framework employs 176 distinct perturbation strategies across six categories of informational uncertainty and three intensity levels. Evaluation of over 30 models uncovers systematic failure patterns: refusal comprises separable detection and categorization skills, and neither scale nor extended reasoning improves performance. We find that selective refusal is a trainable, alignment-sensitive capability, offering a clear path for improvement. We release two benchmarks -- RefusalBench-NQ (single document) and RefusalBench-GaRAGe (multi-document) -- and our complete generation framework to enable continued, dynamic evaluation of this critical capability.

Refusal is a key safety behavior in aligned language models, yet the internal mechanisms driving refusals remain opaque. In this work, we conduct a mechanistic study of refusal in instruction-tuned LLMs using sparse autoencoders to identify latent features that causally mediate refusal behaviors. We apply our method to two open-source chat models and intervene on refusal-related features to assess their influence on generation, validating their behavioral impact across multiple harmful datasets. This enables a fine-grained inspection of how refusal manifests at the activation level and addresses key research questions such as investigating upstream-downstream latent relationship and understanding the mechanisms of adversarial jailbreaking techniques. We also establish the usefulness of refusal features in enhancing generalization for linear probes to out-of-distribution adversarial samples in classification tasks. We open source our code in https://github.com/wj210/refusal_sae.

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.

Refusal behavior in large language models (LLMs) enables them to decline responding to harmful, unethical, or inappropriate prompts, ensuring alignment with ethical standards. This paper investigates refusal behavior across six LLMs from three architectural families. We challenge the assumption of refusal as a linear phenomenon by employing dimensionality reduction techniques, including PCA, t-SNE, and UMAP. Our results reveal that refusal mechanisms exhibit nonlinear, multidimensional characteristics that vary by model architecture and layer. These findings highlight the need for nonlinear interpretability to improve alignment research and inform safer AI deployment strategies.

Arabic large language models must refuse harmful prompts without over-refusing benign or sensitive prompts, yet a single refusal rate hides this trade-off. We evaluate it using benign refusal B and harmful-prompt refusal H, where H measures refusal rather than harmful compliance. Across five Arabic-capable models and 130 runs on the full human-written AraSafe set, refusal-only supervised fine-tuning (SFT) collapses toward blanket refusal, whereas selected mixed-SFT configurations reach H = 90% to 93% at B = 14% to 23%; four selected configurations exceed the H = 90% target in all three runs, while Fanar does so in two of three. Direct Preference Optimization (DPO) and inference guards change B and H differently across models rather than acting as uniform upgrades. In a blinded 300-response audit, annotator binary-refusal agreement is 89.0% (kappa = 0.78); Qwen3Guard and Aya Expanse 32B reach 88.7% and 91.0% accuracy, respectively, with no conclusive paired difference. Selected SFT raises H on Arabizi for all five models, but none reaches 90%, showing only partial transfer from Modern Standard Arabic. Overall, the results support model-specific operating-point selection: set a deployment target and retain only interventions that improve it.

Large Language Models (LLMs) encode behaviors such as refusal within their activation space, yet identifying these behaviors remains a significant challenge. Existing methods often rely on predefined refusal templates detectable in output tokens or require manual analysis. We introduce COSMIC (Cosine Similarity Metrics for Inversion of Concepts), an automated framework for direction selection that identifies viable steering directions and target layers using cosine similarity - entirely independent of model outputs. COSMIC achieves steering performance comparable to prior methods without requiring assumptions about a model's refusal behavior, such as the presence of specific refusal tokens. It reliably identifies refusal directions in adversarial settings and weakly aligned models, and is capable of steering such models toward safer behavior with minimal increase in false refusals, demonstrating robustness across a wide range of alignment conditions.

Refusals - instances where large language models (LLMs) decline or fail to fully execute user instructions - are crucial for both AI safety and AI capabilities and the reduction of hallucinations in particular. These behaviors are learned during post-training, especially in instruction fine-tuning (IFT) and reinforcement learning from human feedback (RLHF). However, existing taxonomies and evaluation datasets for refusals are inadequate, often focusing solely on should-not-related (instead of cannot-related) categories, and lacking tools for auditing refusal content in black-box LLM outputs. We present a comprehensive framework for classifying LLM refusals: (a) a taxonomy of 16 refusal categories, (b) a human-annotated dataset of over 8,600 instances from publicly available IFT and RLHF datasets, (c) a synthetic dataset with 8,000 examples for each refusal category, and (d) classifiers trained for refusal classification. Our work enables precise auditing of refusal behaviors in black-box LLMs and automatic analyses of refusal patterns in large IFT and RLHF datasets. This facilitates the strategic adjustment of LLM refusals, contributing to the development of more safe and reliable LLMs.

LLMs are trained to refuse harmful instructions, but do they truly understand harmfulness beyond just refusing? Prior work has shown that LLMs' refusal behaviors can be mediated by a one-dimensional subspace, i.e., a refusal direction. In this work, we identify a new dimension to analyze safety mechanisms in LLMs, i.e., harmfulness, which is encoded internally as a separate concept from refusal. There exists a harmfulness direction that is distinct from the refusal direction. As causal evidence, steering along the harmfulness direction can lead LLMs to interpret harmless instructions as harmful, but steering along the refusal direction tends to elicit refusal responses directly without reversing the model's judgment on harmfulness. Furthermore, using our identified harmfulness concept, we find that certain jailbreak methods work by reducing the refusal signals without reversing the model's internal belief of harmfulness. We also find that adversarially finetuning models to accept harmful instructions has minimal impact on the model's internal belief of harmfulness. These insights lead to a practical safety application: The model's latent harmfulness representation can serve as an intrinsic safeguard (Latent Guard) for detecting unsafe inputs and reducing over-refusals that is robust to finetuning attacks. For instance, our Latent Guard achieves performance comparable to or better than Llama Guard 3 8B, a dedicated finetuned safeguard model, across different jailbreak methods. Our findings suggest that LLMs' internal understanding of harmfulness is more robust than their refusal decision to diverse input instructions, offering a new perspective to study AI safety

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.

Large language models (LLMs) are typically aligned to comply with safety guidelines by refusing harmful instructions. A recent attack, termed abliteration, isolates and suppresses the single latent direction most responsible for refusal behavior, enabling the model to generate unethical content. We propose a defense that modifies how models generate refusals. We construct an extended-refusal dataset that contains harmful prompts with a full response that justifies the reason for refusal. We then fine-tune Llama-2-7B-Chat and Qwen2.5-Instruct (1.5B and 3B parameters) on our extended-refusal dataset, and evaluate the resulting systems on a set of harmful prompts. In our experiments, extended-refusal models maintain high refusal rates, dropping at most by 10%, whereas baseline models' refusal rates drop by 70-80% after abliteration. A broad evaluation of safety and utility shows that extended-refusal fine-tuning neutralizes the abliteration attack while preserving general performance.

Striking a balance between helpfulness and safety remains a fundamental challenge in aligning large language models. To achieve this balance, models should refuse harmful queries (e.g., "How do I shoot someone?") while remaining responsive to benign inputs, even those superficially resembling harmful queries (e.g., "Where can I shoot a good photo?"). However, models often struggle to distinguish genuinely harmful queries from benign queries that contain superficially risky language, resulting in false refusals. In this paper, we address the issue by decomposing a response in the safety-tuning dataset into two distinct components: (i) a boilerplate refusal statement and (ii) a rationale explaining the refusal. Our experiments and analyses show that refusal statements impede accurate discrimination between harmful and benign queries by inducing reliance on superficial cues. In contrast, training solely on rationales reduces false refusals while maintaining a comparable level of safety performance. Rationale-Only benefits also appear in our ICL configuration and remain compatible with the evaluated inference-time mitigation methods. The results emphasize the necessity of precisely curated, fine-grained safety supervision datasets and outline directions for constructing aligned agents that better reconcile helpfulness with safety.

A key component of building safe and reliable language models is enabling the models to appropriately refuse to follow certain instructions or answer certain questions. We may want models to output refusal messages for various categories of user queries, for example, ill-posed questions, instructions for committing illegal acts, or queries which require information past the model's knowledge horizon. Engineering models that refuse to answer such questions is complicated by the fact that an individual may want their model to exhibit varying levels of sensitivity for refusing queries of various categories, and different users may want different refusal rates. The current default approach involves training multiple models with varying proportions of refusal messages from each category to achieve the desired refusal rates, which is computationally expensive and may require training a new model to accommodate each user's desired preference over refusal rates. To address these challenges, we propose refusal tokens, one such token for each refusal category or a single refusal token, which are prepended to the model's responses during training. We then show how to increase or decrease the probability of generating the refusal token for each category during inference to steer the model's refusal behavior. Refusal tokens enable controlling a single model's refusal rates without the need of any further fine-tuning, but only by selectively intervening during generation.

Safety-aligned language models systematically refuse harmful requests. While activation steering can modulate refusal, ablating the raw "refusal vector" calculated from contrastive harmful and harmless prompts often causes collateral damage and distribution drift. We argue this degradation occurs because the raw vector is polysemantic, entangling the refusal signal with core capability circuits and linguistic style. We introduce Surgical Refusal Ablation (SRA) to distill these steering directions. SRA constructs a registry of independent Concept Atoms representing protected capabilities and stylistic confounds, then uses ridge-regularized spectral residualization to orthogonalize the refusal vector against these directions. This yields a clean refusal direction that targets refusal-relevant structure while minimizing disruption to the model's semantic geometry. Across five models (Qwen3-VL and Ministral series), SRA achieves deep refusal reduction (0-2%) with negligible perplexity impact on Wikitext-2 (mean delta PPL approx. 0.02) and minimal distribution drift. Notably, standard ablation on Qwen3-VL-4B induces severe drift (first-token KL = 2.088), whereas SRA maintains the original distribution (KL = 0.044) while achieving the same 0% refusal rate. Using teacher-forced perplexity on GSM8K and MBPP as a high-resolution capability proxy, we show SRA preserves math and code distributions. These results suggest that common "model damage" is often "Ghost Noise," defined as the spectral bleeding of the dirty refusal direction into capability subspaces.

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.

Modern deployments require LLMs to enforce safety policies at scale, yet many controls rely on inference-time interventions that add recurring compute cost and serving complexity. Activation steering is widely used, but it requires runtime hooks and scales cost with the number of generations; conditional variants improve selectivity by gating when steering is applied but still retain an inference-time control path. We ask whether selective refusal can be moved entirely offline: can a mechanistic understanding of category-specific refusal be distilled into a circuit-restricted weight update that deploys as a standard checkpoint? We propose C-Δθ: Circuit Restricted Weight Arithmetic, which (i) localizes refusal-causal computation as a sparse circuit using EAP-IG and (ii) computes a constrained weight update ΔθC supported only on that circuit (typically <5% of parameters). Applying ΔθC yields a drop-in edited checkpoint with no inference-time hooks, shifting cost from per-request intervention to a one-time offline update. We evaluate category-targeted selectivity and capability retention on refusal and utility benchmarks.

When a language model refuses to answer a prompt, it is unclear whether the correct answer is erased from its internal representations, or merely suppressed at the output layer. We investigate this mechanism using a controlled withhold setting, which yields perfectly matched answering and refusal trajectories for bidirectional activation patching. We uncover a causal asymmetry in intervention locality under matched causal interventions, which we term broken symmetry. Even when a model generates a clean refusal, the correct answer remains linearly recoverable from its hidden states. Furthermore, releasing this withheld answer is a highly local operation, requiring only a single-position patch. Conversely, the reverse operation is not equally local: reimposing suppression requires broader interventions across multiple positions, and assembling a coherent refusal sequence is more difficult still. We further demonstrate that while an average answer-to-refusal displacement vector marks the geometric difference between these states, it fails to act as a reliable, reversible linear control toggle between behaviours. Taken together, our findings show that refusal does not function as a simple symmetric switch. For safety and auditing, this implies that probe recoverability can overestimate true behavioural control, and locating refusal-relevant directions does not reliably grant the ability to steer a model from answering to coherent refusal.

Mechanistic interpretability has emerged as the primary means to understand safety behavior of LLMs. However, existing tools primarily focus on the activating neurons or features of a model. The role of the remaining large set of inactive components is invisible to such methods. This work demonstrates that the inactive set contains safety-critical features that are causally relevant for refusal of harmful prompts. Suppressing such features could turn refusals into compliance, while passing undetected by prevalent interpretability tools. We introduce the Counterfactual Activation Potential (CAP), a metric that quantifies a suppressed feature's latent activation tendency as the product of its encoder alignment (how strongly the input drives it), suppression strength (how strongly active features inhibit it), and safety criticality (how much refusal depends on it). To find suppressed safety features at scale, we propose CAP-guided Safety Feature Discovery (CSFD), a two-stage filtering algorithm that identifies candidate safety features from hundreds of thousands of transcoder features without exhaustive ablation. A significant fraction of trials turn compliant with harmful prompts when a candidate feature is ablated. Under natural jailbreaks, the suppression acting on the highest-CAP features rises 2-4x, and their activation correspondingly falls by up to 80%. Amplifying a feature's suppressors pushes its activation down and raises harmful compliance with prompts related to the suppressed feature, with no such effect for random features. Our experiments span five Gemma, Qwen, and Llama models across various parameter sizes. Our findings indicate that jailbreaks could operate in part by suppressing safety-critical features rather than solely activating harmful ones, and that suppressed features are a necessary complement to activation-focused interpretability of safety behavior.

Leaderboards rank long-horizon agents by their final outputs. Yet a higher score alone does not establish whether two systems are comparable or which stage accounts for the difference. Unequal evidence, inputs, or budgets can affect scores, and statistical corrections do not remove this mismatch. We introduce compositional controllability to address these questions. A comparison window covers one stage, several stages, or the whole agent. Our central result bounds the gap between observed and controlled score differences using only nuisance outside the window. This yields an admissibility test applied before scores are inspected. Inadmissible comparisons are refused. For admissible pairs, an ordering is certified only when the score gap exceeds the combined sampling and nuisance radii; otherwise, it remains undecided. These decisions give each system a rank interval. We introduce BioLitBench, a benchmark of 2,042 biomedical articles represented as structured claim graphs. Among seven published pipelines, a conventional statistical analysis declares a winner in 14 of 21 pairwise comparisons. Yet the top-ranked system alone received the target review's bibliography. To isolate pipeline performance, our test requires matched inputs and a fixed backbone model. It refuses 11 of the 21 comparisons, including every comparison involving the top-ranked system. Seven of the 14 conventional conclusions fall within these refused pairs. The same comparison windows support stage-level training. We train SCRIBE on Qwen3.8-27B using rewards measured at each stage's exit. Under matched evidence, SCRIBE achieves a certified rank interval of [1,2], with certified advantages over all evaluated published pipelines and the evaluated Claude and OpenAI agents. Under same pool, SCRIBE matches the strongest published retriever and is certified above three published pipelines.

Prior work argues that refusal in large language models is mediated by a single activation-space direction, enabling effective steering and ablation. We show that this account is incomplete. Across eleven categories of refusal and non-compliance, including safety, incomplete or unsupported requests, anthropomorphization, and over-refusal, we find that these refusal behaviors correspond to geometrically distinct directions in activation space. Yet despite this diversity, linear steering along any refusal-related direction produces nearly identical refusal to over-refusal trade-offs, acting as a shared one-dimensional control knob. The primary effect of different directions is not whether the model refuses, but how it refuses.

We introduce Refusal Steering, an inference-time method to exercise fine-grained control over Large Language Models refusal behaviour on politically sensitive topics without retraining. We replace fragile pattern-based refusal detection with an LLM-as-a-judge that assigns refusal confidence scores and we propose a ridge-regularized variant to compute steering vectors that better isolate the refusal--compliance direction. On Qwen3-Next-80B-A3B-Thinking, our method removes the refusal behaviour of the model around politically sensitive topics while maintaining safety on JailbreakBench and near-baseline performance on general benchmarks. The approach generalizes across 4B and 80B models and can also induce targeted refusals when desired. We analize the steering vectors and show that refusal signals concentrate in deeper layers of the transformer and are distributed across many dimensions. Together, these results demonstrate that activation steering can remove political refusal behaviour while retaining safety alignment for harmful content, offering a practical path to controllable, transparent moderation at inference time.

The safety alignment of large language models (LLMs) can be circumvented through adversarially crafted inputs, yet the mechanisms by which these attacks bypass safety barriers remain poorly understood. Prior work suggests that a single refusal direction in the model's activation space determines whether an LLM refuses a request. In this study, we propose a novel gradient-based approach to representation engineering and use it to identify refusal directions. Contrary to prior work, we uncover multiple independent directions and even multi-dimensional concept cones that mediate refusal. Moreover, we show that orthogonality alone does not imply independence under intervention, motivating the notion of representational independence that accounts for both linear and non-linear effects. Using this framework, we identify mechanistically independent refusal directions. We show that refusal mechanisms in LLMs are governed by complex spatial structures and identify functionally independent directions, confirming that multiple distinct mechanisms drive refusal behavior. Our gradient-based approach uncovers these mechanisms and can further serve as a foundation for future work on understanding LLMs.

Large language models often answer structurally unanswerable questions, such as computing cot(-540°) or evaluating (1).startswith("1"), instead of abstaining. We ask whether this failure reflects missing recognition or failed routing from recognition to abstention. Across instruction-tuned models from 1.7B to 70B parameters, a single linear direction in the hidden state separates answerable from structurally impossible math and code prompts, showing that models represent impossibility before generation. Yet this recognition direction is nearly orthogonal to the canonical safety-refusal direction that mediates trained harmful-content refusal. An in-domain behavior-defined invalidity-aware direction is closer to recognition, but only partially aligned with it, and remains near-orthogonal to safety refusal. Generation-time steering along the recognition direction changes invalidity-aware behavior bidirectionally and dose-responsively on structural math and code cells, while random directions do not. Base/instruct comparisons further show that the low-cosine geometry is already present at the pretraining endpoint. The confident-on-impossible failure is therefore better explained as a routing failure than as an encoding failure: the model has a usable "no admissible answer" signal, but the safety-refusal pathway is not aligned to use it.

Refusal mechanisms in large language models (LLMs) are essential for ensuring safety. Recent research has revealed that refusal behavior can be mediated by a single direction in activation space, enabling targeted interventions to bypass refusals. While this is primarily demonstrated in an English-centric context, appropriate refusal behavior is important for any language, but poorly understood. In this paper, we investigate the refusal behavior in LLMs across 14 languages using PolyRefuse, a multilingual safety dataset created by translating malicious and benign English prompts into these languages. We uncover the surprising cross-lingual universality of the refusal direction: a vector extracted from English can bypass refusals in other languages with near-perfect effectiveness, without any additional fine-tuning. Even more remarkably, refusal directions derived from any safety-aligned language transfer seamlessly to others. We attribute this transferability to the parallelism of refusal vectors across languages in the embedding space and identify the underlying mechanism behind cross-lingual jailbreaks. These findings provide actionable insights for building more robust multilingual safety defenses and pave the way for a deeper mechanistic understanding of cross-lingual vulnerabilities in LLMs.

Biosecurity evaluations of language models typically ask whether models produce hazardous output. This paper asks a complementary question: when a model refuses, is that refusal structurally sound, or does it disappear under modest changes to prompt framing, formatting, or output length? Across five architectures, no model cleanly discriminated benign from hazard. Gemma 2 2B-IT never genuinely refused across 75 prompts, hedging on every hazard-adjacent query. Gemma 4 E2B-IT refused 65/75 prompts with chat-template formatting and 0/75 without it. Both Gemma models collapsed to 0% under an 80-token cap. Qwen 2.5 1.5B and Phi-3-mini over-refused, flagging 83-87% of benign biology as hazardous. Llama 3.2 1B showed the only meaningful tier gradient (61-point spread). To probe what drives such over-refusal, we tested a panel of Schedule I but biologically non-toxic compounds (notably psilocybin cultivation, with FDA Breakthrough Therapy status). Some models refused these at rates exceeding genuinely hazardous biology, suggesting refusal tracks legality and cultural salience over CBRN hazard. To measure the internal side, we introduce a divergence score D comparing a model's surface response label to its internal sparse autoencoder (SAE) feature activations. Full D was computed on Gemma 2 2B-IT (Gemma Scope 1) and Gemma 4 E2B-IT (author-trained bio SAE). Two fine-tuned Gemma 2 domain SAEs were released. On Gemma 4, comply and refuse responses separated by a 0.647-point gap with zero overlap (n=75), though this is preliminary, with a narrow catalog, within-sample calibration, and Gemma-family-only SAE coverage. Built over one hackathon weekend on consumer hardware (GTX 1650 Ti Max-Q, plus Colab T4 for SAE training), this preliminary evidence suggests activation-level auditing may surface failure modes invisible to behavioral evaluation, with substantial variation across architectures.

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.

Refusal on harmful prompts is a key safety behaviour in instruction-tuned large language models (LLMs), yet the internal causes of this behaviour remain poorly understood. We study two public instruction-tuned models, Gemma-2-2B-IT and LLaMA-3.1-8B-IT, using sparse autoencoders (SAEs) trained on residual-stream activations. Given a harmful prompt, we search the SAE latent space for feature sets whose ablation flips the model from refusal to compliance, demonstrating causal influence and creating a jailbreak. Our search proceeds in three stages: (1) Refusal Direction: find a refusal-mediating direction and collect SAE features near that direction; (2) Greedy Filtering: prune to a minimal set; and (3) Interaction Discovery: fit a factorization machine (FM) that captures nonlinear interactions among the remaining active features and the minimal set. This pipeline yields a broad set of jailbreak-critical features, offering insight into the mechanistic basis of refusal. Moreover, we find evidence of redundant features that remain dormant unless earlier features are suppressed. Our findings highlight the potential for fine-grained auditing and targeted intervention in safety behaviours by manipulating the interpretable latent space.

Large Language Models (LLMs) should refuse to answer questions beyond their knowledge. This capability, which we term knowledge-aware refusal, is crucial for factual reliability. However, existing metrics fail to faithfully measure this ability. On the one hand, simple refusal-based metrics are biased by refusal rates and yield inconsistent scores when models exhibit different refusal tendencies. On the other hand, existing calibration metrics are proxy-based, capturing the performance of auxiliary calibration processes rather than the model's actual refusal behavior. In this work, we propose the Refusal Index (RI), a principled metric that measures how accurately LLMs refuse questions they do not know. We define RI as Spearman's rank correlation between refusal probability and error probability. To make RI practically measurable, we design a lightweight two-pass evaluation method that efficiently estimates RI from observed refusal rates across two standard evaluation runs. Extensive experiments across 16 models and 5 datasets demonstrate that RI accurately quantifies a model's intrinsic knowledge-aware refusal capability in factual tasks. Notably, RI remains stable across different refusal rates and provides consistent model rankings independent of a model's overall accuracy and refusal rates. More importantly, RI provides insight into an important but previously overlooked aspect of LLM factuality: while LLMs achieve high accuracy on factual tasks, their refusal behavior can be unreliable and fragile. This finding highlights the need to complement traditional accuracy metrics with the Refusal Index for comprehensive factuality evaluation.

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-aligned large language models (LLMs) remain vulnerable to jailbreaks and can over-refuse benign requests. Understanding these failures requires distinguishing harmfulness recognition from the computation that writes refusal. We reframe this as a pathway question and make it measurable on residual-stream updates in decoder-only Transformers, separating the harmfulness information a module update exposes from the change it produces in readable refusal evidence. Across 18 Base-Instruct pairs spanning Qwen, Llama, and Gemma, harmfulness is decoded more readily from attention updates, particularly in early layers, whereas MLP updates contribute more strongly to harm-selective refusal writing in aligned models. Post-training strengthens MLP refusal writing in all 18 pairs without systematically improving harmfulness decodability, while positive MLP writing becomes more concentrated in deeper layers after post-training. Across six reasoning-enabled models, harmfulness remains decodable during reasoning while strong refusal writing concentrates at answer onset, a pattern that persists with reasoning disabled. Causal interventions further support this pathway: resetting downstream MLP updates to their clean values substantially reduces the refusal shift caused by upstream attention interventions. Building on these findings, we introduce Sentinel, which combines an attention-derived perception score P with an MLP-derived decision score D. Their joint profiles characterize jailbreaks and over-refusals, and conditional routing with both signals achieves a lower aggregate error rate than single-signal routing and evaluated fixed reminders across eight models.

Current evaluations of multimodal unlearning typically focus on whether a model stops outputting target information. This surface-level criterion, however, masks two critical failure modes. First, knowledge suppression techniques often induce confabulation, leading the model to hallucinate facts about the unlearned subject. Second, refusal training merely alters observable behavior without erasing the underlying knowledge; such models might decline open queries but still identify correct answers under multiple-choice probing. To overcome these limitations, we propose Forget-Aware Self-Distillation (FASD). This method is designed to achieve honest ignorance, a state where the model genuinely loses the target knowledge and learns to acknowledge its absence. FASD separates these goals into two complementary channels. The behavior channel distills appropriate refusals from an ignorance-prompted copy of the model itself. Crucially, target tokens are masked from the teacher distribution to avoid accidentally reteaching the forgotten concepts. Simultaneously, the knowledge channel directly suppresses ground-truth answer spans independently of the model's generated responses. Experiments demonstrate that FASD successfully unlearns target knowledge, prevents information leakage in open-ended tasks, and preserves performance on retained data. The resulting model not only forgets the specified information but accurately recognizes the boundaries of its own knowledge. Code and data will be released upon publication.

Machine learning models always make a prediction, even when it is likely to be inaccurate. This behavior should be avoided in many decision support applications, where mistakes can have severe consequences. Albeit already studied in 1970, machine learning with rejection recently gained interest. This machine learning subfield enables machine learning models to abstain from making a prediction when likely to make a mistake. This survey aims to provide an overview on machine learning with rejection. We introduce the conditions leading to two types of rejection, ambiguity and novelty rejection, which we carefully formalize. Moreover, we review and categorize strategies to evaluate a model's predictive and rejective quality. Additionally, we define the existing architectures for models with rejection and describe the standard techniques for learning such models. Finally, we provide examples of relevant application domains and show how machine learning with rejection relates to other machine learning research areas.

Recent unlearning methods (e.g. NPO, DPO, LUNAR) make use of refusal alignment to suppress forgotten data. However, it has been shown that refusal responses might leave traces of unlearning, and recent attacks have been able to successfully recover some of the unlearned knowledge. In this paper, we uncover a new vulnerability. Existing attacks typically assume that the forgotten prompts are already known to the adversary and focus on recovering their answers. However, we show that the forgotten prompts themselves can be extracted by using the retained data and black-box access to the model. Our attack, Targeted Active Search (TAS), first identifies the forgotten entities by constructing canonical templates and entity pool, and selectively querying the model using the most informative template-entity pair under a limited query budget. Once the entities are identified, TAS instantiates prompt templates with those entities to probe the unlearned model and reconstruct the forgotten prompts. Experiments across three unlearning methods with three datasets and three LLMs shows that TAS recovers the forgotten entity with 100% accuracy and reconstructs up to 95% of forgotten prompts, all while using up to 99.7% fewer queries than naive probing.

This study reveals a previously unexplored vulnerability in the safety alignment of Large Language Models (LLMs). Existing aligned LLMs predominantly respond to unsafe queries with refusals, which often begin with a fixed set of prefixes (I'm sorry). We demonstrate that this rigid refusal pattern is a vulnerability and introduce a novel refusal unlearning technique that exploits it. Specifically, we fine-tune LLMs using merely 1,000 benign samples, where each response is prepended with a refusal prefix. The underlying intuition is to disrupt the refusal completion pathway, thereby driving the model to forget how to refuse while following harmful instructions. This intuition is further supported by theoretical proofs. We apply this approach to a total of 16 LLMs, including various open-source models from Llama, Qwen, and Gemma families, as well as closed-source models such as Gemini and GPT. Experimental results show that the safety scores of previously aligned LLMs degrade both consistently and substantially. Importantly, we verify that the observed gain cannot be attributed to plain fine-tuning or random prefix effects. Our findings suggest that current safety alignment may rely heavily on token sequence memorization rather than reasoning, motivating future work beyond simple refusal mechanisms. Code has been released: https://github.com/guoyang9/refusal-unlearning.

Whether large language models can be prompted to generate rhetorical fallacies on demand, and whether current safety post-training constrains this behavior, has received less attention than the related question of detecting fallacies in existing text. We close this gap with DeflectBench, evaluating 23,990 generations from four frontier models across three deflection strategies (whataboutism, ad hominem, red herring), seven prompt framings, and 80 claims spanning four controversy levels. Refusal is governed primarily by request structure rather than claim content. Per claim refusal varies by only 11 percentage points across the 80 claims, while a single prompt frame change can swing within model refusal by nearly 100 percentage points and switching the requested fallacy type can swing it by over 80 percentage points within explicit framings. An educational debate coach prompt framing collapses refusal to near zero across all four model families, but the bypassed behavior is not clean compliance. Models typically produce labeled compliance, naming the requested manipulation in the same response that contains it. The four models distribute differently across refusal, labeled compliance, soft refusal, and clean compliance. The code and dataset are released at https://github.com/ArtKanke/DeflectBench.

Current safety alignment of foundation models largely follows a one-size-fits-all paradigm, applying the same refusal policy across users and contexts. As a result, models may refuse requests that are unsafe for general users but legitimate for authorized professionals, limiting helpfulness in specialized professional settings. Existing approaches either require costly realignment or rely on inference-time steering that suffers from imprecise control and added latency. To this end, we propose Palette, a modular, controllable, and efficient framework that selectively relaxes refusal behavior on authorized target domains while preserving standard safety elsewhere. Our method identifies a refusal direction via multi-objective search and internalizes it into the model through lightweight adaptation. Palette further supports modular composition: it learns domain-specific safety controls independently and composes them through parameter merging, enabling on-demand multi-domain authorization without retraining. Experiments across four safety benchmarks, multiple model variants, and both LLMs and VLMs show that Palette delivers precise safety control without sacrificing general utility, offering a practical path toward foundation models that adapt to diverse professional needs.

Safety alignment instills in Large Language Models (LLMs) a critical capacity to refuse malicious requests. Prior works have modeled this refusal mechanism as a single linear direction in the activation space. We posit that this is an oversimplification that conflates two functionally distinct neural processes: the detection of harm and the execution of a refusal. In this work, we deconstruct this single representation into a Harm Detection Direction and a Refusal Execution Direction. Leveraging this fine-grained model, we introduce Differentiated Bi-Directional Intervention (DBDI), a new white-box framework that precisely neutralizes the safety alignment at critical layer. DBDI applies adaptive projection nullification to the refusal execution direction while suppressing the harm detection direction via direct steering. Extensive experiments demonstrate that DBDI outperforms prominent jailbreaking methods, achieving up to a 97.88\% attack success rate on models such as Llama-2. By providing a more granular and mechanistic framework, our work offers a new direction for the in-depth understanding of LLM safety alignment.

Since the release of OpenAI's ChatGPT, generative language models have attracted extensive public attention. The increased usage has highlighted generative models' broad utility, but also revealed several forms of embedded bias. Some is induced by the pre-training corpus; but additional bias specific to generative models arises from the use of subjective fine-tuning to avoid generating harmful content. Fine-tuning bias may come from individual engineers and company policies, and affects which prompts the model chooses to refuse. In this experiment, we characterize ChatGPT's refusal behavior using a black-box attack. We first query ChatGPT with a variety of offensive and benign prompts (n=1,706), then manually label each response as compliance or refusal. Manual examination of responses reveals that refusal is not cleanly binary, and lies on a continuum; as such, we map several different kinds of responses to a binary of compliance or refusal. The small manually-labeled dataset is used to train a refusal classifier, which achieves an accuracy of 96%. Second, we use this refusal classifier to bootstrap a larger (n=10,000) dataset adapted from the Quora Insincere Questions dataset. With this machine-labeled data, we train a prompt classifier to predict whether ChatGPT will refuse a given question, without seeing ChatGPT's response. This prompt classifier achieves 76% accuracy on a test set of manually labeled questions (n=985). We examine our classifiers and the prompt n-grams that are most predictive of either compliance or refusal. Our datasets and code are available at https://github.com/maxwellreuter/chatgpt-refusals.

LLMs increasingly exhibit over-refusal behavior, where safety mechanisms cause models to reject benign instructions that superficially resemble harmful content. This phenomena diminishes utility in production applications that repeatedly rely on common prompt templates or applications that frequently rely on LLMs for specific tasks (e.g. sentiment analysis, language translation). Through comprehensive evaluation, we demonstrate that LLMs still tend to refuse responses to harmful instructions when those instructions are reframed to appear as benign tasks. Our mechanistic analysis reveal that LLMs follow distinct "constellation" patterns in embedding space as representations traverse layers, with each task maintaining consistent trajectories that shift predictably between refusal and non-refusal cases. We introduce SafeConstellations, an inference-time trajectory-shifting approach that tracks task-specific trajectory patterns and guides representations toward non-refusal pathways. By selectively guiding model behavior only on tasks prone to over-refusal, and by preserving general model behavior, our method reduces over-refusal rates by up to 73% with minimal impact on utility-offering a principled approach to mitigating over-refusals.

Reliable Large Language Models (LLMs) should abstain when confidence is insufficient. However, prior studies often treat refusal as a generic "I don't know'', failing to distinguish input-level ambiguity (data uncertainty) from capability limitations (model uncertainty). This lack of distinction limits downstream action decisions like requesting clarification or invoking external tools. In this work, we introduce UA-Bench, a benchmark of over 3,500 questions drawn from six datasets spanning knowledge-intensive and reasoning-intensive tasks, designed to evaluate explicit uncertainty attribution. An evaluation of 18 frontier LLMs shows that even state-of-the-art models struggle to reliably discriminate between data uncertainty and model uncertainty, and that high answer accuracy does not necessarily imply strong uncertainty attribution ability. To narrow this gap, we propose a lightweight data synthesis and reinforcement learning strategy. Experiments on both Qwen3-4B-Instruct-2507 and Qwen3-8B in thinking mode show that the proposed method improves uncertainty attribution while preserving answer accuracy. Our code and data are publicly available now.

We identify a structural weakness in current large language model (LLM) alignment: modern refusal mechanisms are fail-open. While existing approaches encode refusal behaviors across multiple latent features, suppressing a single dominant feature-via prompt-based jailbreaks-can cause alignment to collapse, leading to unsafe generation. Motivated by this, we propose fail-closed alignment as a design principle for robust LLM safety: refusal mechanisms should remain effective even under partial failures via redundant, independent causal pathways. We present a concrete instantiation of this principle: a progressive alignment framework that iteratively identifies and ablates previously learned refusal directions, forcing the model to reconstruct safety along new, independent subspaces. Across four jailbreak attacks, we achieve the strongest overall robustness while mitigating over-refusal and preserving generation quality, with small computational overhead. Our mechanistic analyses confirm that models trained with our method encode multiple, causally independent refusal directions that prompt-based jailbreaks cannot suppress simultaneously, providing empirical support for fail-closed alignment as a principled foundation for robust LLM safety.

Large language models (LLMs) have transformed the way we access information. These models are often tuned to refuse to comply with requests that are considered harmful and to produce responses that better align with the preferences of those who control the models. To understand how this "censorship" works. We use representation engineering techniques to study open-weights safety-tuned models. We present a method for finding a refusal--compliance vector that detects and controls the level of censorship in model outputs. We also analyze recent reasoning LLMs, distilled from DeepSeek-R1, and uncover an additional dimension of censorship through "thought suppression". We show a similar approach can be used to find a vector that suppresses the model's reasoning process, allowing us to remove censorship by applying the negative multiples of this vector

Large language models (LLMs) routinely face requests that should be refused, creating a trade-off between helpfulness and harm prevention. However, refusals themselves can be helpful. In high-risk interactions involving crisis, coercion, or escalating intent, blunt non-compliance may prevent direct harm while still failing to support the needs of the person behind the request. We present PsychoSafe, a psychologically-informed refusal framework that reframes refusal as structured supportive communication grounded in evidence-based intervention strategies. To develop PsychoSafe, we construct a corpus of 8019 prompt-response pairs spanning five psychologically salient risk domains and apply prompting and parameter-efficient fine-tuning to Qwen 3.5 27B. On a balanced validation set of 500 prompts, evaluated with an LLM judge and validated through human ratings, PsychoSafe prompting improves overall refusal quality by 28.1% over a generic baseline, with particularly strong gains in external resource referral (+46.8%) and psychological grounding (+34.8%), while preserving downstream performance on non-refusal tasks. Fine-tuning achieves near-perfect refusal and resource-referral rates but reduces response relevance. Additional evaluations on SORRY-Bench and XSTest show strong in-domain robustness but limited out-of-domain generalization, suggesting that future work should diversify fine-tuning data to help models apply interventions selectively rather than schematically.

Large Language Models used in ChatGPT have traditionally been trained to learn a refusal boundary: depending on the user's intent, the model is taught to either fully comply or outright refuse. While this is a strong mitigation for explicitly malicious prompts, focusing safety training on refusals can lead to brittleness for prompts with obscured user intent. Binary refusal boundaries are especially ill-suited for dual-use cases (such as biology or cybersecurity), where a user request can be answered safely at a high level, but in some cases can lead to malicious uplift if sufficiently detailed or actionable. As an alternative, we propose safe-completions: a safety-training approach that centers on the safety of the assistant's output, rather than a binary classification of the user's intent. Safe-completions seek to maximize helpfulness within the safety policy's constraints. We incorporated this approach into GPT-5 and find that across both production comparisons and internally controlled experiments, safe-completion training improves safety (especially on dual-use prompts), reduces the severity of residual safety failures, and substantially increases model helpfulness.

Safety-aligned large language models (LLMs) sometimes falsely refuse pseudo-harmful prompts, like "how to kill a mosquito," which are actually harmless. Frequent false refusals not only frustrate users but also provoke a public backlash against the very values alignment seeks to protect. In this paper, we propose the first method to auto-generate diverse, content-controlled, and model-dependent pseudo-harmful prompts. Using this method, we construct an evaluation dataset called PHTest, which is ten times larger than existing datasets, covers more false refusal patterns, and separately labels controversial prompts. We evaluate 20 LLMs on PHTest, uncovering new insights due to its scale and labeling. Our findings reveal a trade-off between minimizing false refusals and improving safety against jailbreak attacks. Moreover, we show that many jailbreak defenses significantly increase the false refusal rates, thereby undermining usability. Our method and dataset can help developers evaluate and fine-tune safer and more usable LLMs. Our code and dataset are available at https://github.com/umd-huang-lab/FalseRefusal

Safety alignment in instruction-tuned large language models (LLMs) depends on a model's ability to reliably refuse to respond to harmful or disallowed requests. Recent work has shown that a steering vector can be applied to a dense LLM during inference to effectively suppress refusal behavior, inducing response to harmful requests. We extend this refusal steering method to three open-source Mixture-of-Experts (MoE) LLMs and find that steering performance is uninhibited by the complex routing patterns inherent to the MoE architecture. We then propose two expert-aware refusal steering methods that leverage refusal-specific expert routing patterns and expert-specific steering directions to suppress normal refusal behavior. We find that refusal behavior can be effectively steered based on the output of a single expert. Our results show that refusal signals captured by steering methods differ from expert routing behavior, suggesting a substantial role for attention in MoE refusal behavior.

A general-purpose language model that answers a harmful question returns text; a coding model that complies with a malicious request can return a working weapon -- a keylogger, a ransomware stub, an exploit that runs as written. This asymmetry in the severity of a single act of compliance implies coding-specialized models should clear a higher refusal bar than general-purpose chat models, not a lower one, yet the field cannot presently tell whether they do. Refusal benchmarks for malicious code are fragmented: they mix requests for executable software (ready-to-run weapons) with requests for harmful security knowledge (information a human must still operationalise) and report refusal rates over non-comparable corpora, so no single statistic measures the property that actually matters. This paper introduces an expanded consensus-labeled prompt bank that distinguishes between these two request types and provides a construct-stable substrate for cross-corpus coding-model compliance measurement. Eight corpora (ASTRA, CySecBench, AdvBench/harmful_behaviors, JailbreakBench, MalwareBench, RedCode, RMCBench, Scam2Prompt) are consolidated and classified under a five-judge consensus protocol (6,675 prompts x 5 judges = 33,375 calls). The panel reaches Fleiss' kappa = 0.767 [95% CI 0.755, 0.777] ("substantial"); 95.0% of prompts draw at least four agreeing judges, 76.9% are unanimous, and the panel reproduces the earlier four-corpus release at Cohen's kappa = 0.952 on the 3,133 shared prompts. The released bank comprises 4,748 consensus-CODE prompts (executable malicious code requests) and 1,923 consensus-KNOWLEDGE prompts (harmful security knowledge requests). The bank is the validated instrument the field has lacked: a reliability-quantified basis for testing whether coding models meet the stricter refusal standard their executable output demands.

Fine-tuning adapts aligned large language models (LLMs) to downstream tasks, but a few dozen harmful examples can remove their refusal of harmful requests. Prior work localizes safety-related behavior to specific layers, directions, and tokens, suggesting targets for protection. We test whether successful localization and recovery support defenses that survive changes in the attack. Across six checkpoints from four model families, harmful and benign prompts remain linearly separable after attack, and patching full clean hidden states into the compromised model restores refusal at a reproducible transition depth. Building on a prior layer-freezing defense, we freeze every layer up to this depth and repeat the attack. At a hundred harmful examples, refusal remains near zero on all six checkpoints, with recovery transitions above the frozen boundary. In a second study, removing the update's top two singular directions restores refusal after short attention-only fine-tunes on four checkpoints. On Llama-3.1-8B, ordinary training changes weaken this repair and an attacker who spreads the update defeats it. A spectral detector calibrated on benign Llama fine-tunes misses most repair failures on that checkpoint. Localized freezing can nevertheless help preserve refusal when a few harmful examples enter training data unintentionally. These results show that an attacker can bypass a region identified by recovery and defeat a repair that works across multiple checkpoints, motivating five checks for defenses against adaptive fine-tuning. Code is available at https://github.com/js-lee-AI/refusal-relocates.

Refusal training is widely used to prevent LLMs from generating harmful, undesirable, or illegal outputs. We reveal a curious generalization gap in the current refusal training approaches: simply reformulating a harmful request in the past tense (e.g., "How to make a Molotov cocktail?" to "How did people make a Molotov cocktail?") is often sufficient to jailbreak many state-of-the-art LLMs. We systematically evaluate this method on Llama-3 8B, Claude-3.5 Sonnet, GPT-3.5 Turbo, Gemma-2 9B, Phi-3-Mini, GPT-4o mini, GPT-4o, and R2D2 models using GPT-3.5 Turbo as a reformulation model. For example, the success rate of this simple attack on GPT-4o increases from 1% using direct requests to 88% using 20 past tense reformulation attempts on harmful requests from JailbreakBench with GPT-4 as a jailbreak judge. Interestingly, we also find that reformulations in the future tense are less effective, suggesting that refusal guardrails tend to consider past historical questions more benign than hypothetical future questions. Moreover, our experiments on fine-tuning GPT-3.5 Turbo show that defending against past reformulations is feasible when past tense examples are explicitly included in the fine-tuning data. Overall, our findings highlight that the widely used alignment techniques -- such as SFT, RLHF, and adversarial training -- employed to align the studied models can be brittle and do not always generalize as intended. We provide code and jailbreak artifacts at https://github.com/tml-epfl/llm-past-tense.

Large reasoning models (LRMs) generate chain-of-thought (CoT) traces before producing final outputs, introducing a dynamic internal state that may complicate control mechanisms such as refusal. Unlike instruction-tuned LLMs, where refusal is mediated by a single directional subspace, refusal in large reasoning models (LRMs) additionally depends on the CoT. In DeepSeek-R1-Distill-LLaMA-8B, activation steering reverses refusal in only 39% of cases when the CoT is kept fixed, but removing the CoT entirely increases this to 70%, indicating that the CoT actively reinforces refusal. In a two-stage intervention where the model regenerates its CoT under activation steering, refusal is reversed in 94% of cases, while the resulting CoT alone retains 48% of this effect even after steering is removed. This suggests that the CoT can carry and reconstruct the compliance signal independently. These findings indicate that refusal in LRMs is jointly encoded in residual stream activations and CoT. This joint activation makes LRM more robust against activation-level interventions alone, but exposes CoT to a possible alternative surface attack.

Safety-aligned language models often refuse prompts that are actually harmless. Current evaluations mostly report global rates such as false rejection or compliance. These scores treat each prompt alone and miss local inconsistency, where a model accepts one phrasing of an intent but rejects a close paraphrase. This gap limits diagnosis and tuning. We introduce "semantic confusion," a failure mode that captures such local inconsistency, and a framework to measure it. We build ParaGuard, a 10k-prompt corpus of controlled paraphrase clusters that hold intent fixed while varying surface form. We then propose three model-agnostic metrics at the token level: Confusion Index, Confusion Rate, and Confusion Depth. These metrics compare each refusal to its nearest accepted neighbors and use token embeddings, next-token probabilities, and perplexity signals. Experiments across diverse model families and deployment guards show that global false-rejection rate hides critical structure. Our metrics reveal globally unstable boundaries in some settings, localized pockets of inconsistency in others, and cases where stricter refusal does not increase inconsistency. We also show how confusion-aware auditing separates how often a system refuses from how sensibly it refuses. This gives developers a practical signal to reduce false refusals while preserving safety.

Studying how fine-tuning shapes refusal and noncompliance behaviour requires knowing which training examples refuse or otherwise fail to fulfil the request. Existing annotations cover evaluation sets, which are far smaller than training corpora. We present CompOrca, compliance labels for all 4,233,923 examples of the OpenOrca corpus. Every example was classified as compliant or noncompliant by five passes of an open-weight LLM judge (LongCat-2.0, 1.6T parameters). The corpus is released as unanimous compliance (94.75%), unanimous noncompliance (1.28%), and nonunanimous rows (3.97%), with the raw vote counts. A single pass flags 2.7-3.2% of the corpus as noncompliant, while only 1.28% is flagged by all five, so the most ambiguous rows can be filtered out. Against 450 human-annotated examples (150 annotated twice; human-human κ=0.93), the unanimous compliance and noncompliance labels are 97.3% and 86.7% precise. The noncompliance label is a high-precision subset of the corpus's noncompliance. Published refusal-detection methods recall between 0.4% and 94.1% of human-labelled noncompliance. We release the full corpus with its per-row labels and vote counts at https://huggingface.co/datasets/cemiu/CompOrca

Directional ablation removes an aligned language model's ability to refuse by projecting a single "refusal direction" out of the weights that write the residual stream. It needs no gradient-based training and no optimization, only a few hundred contrastive prompts, which makes it the canonical white-box attack on open-weight alignment. However, it has been established only on dense models up to roughly 70B parameters. We study whether it survives the shift to frontier mixture-of-experts (MoE) models whose residual streams are no longer a single tensor and whose weights ship quantized. We apply it to GLM-5.3-Flash (320B parameters, 288 routed experts, a four-wide hyper-connection residual, block-FP8). The attack survives the architecture, but what it reaches is no longer where a reader of the original recipe would look for it. Editing the attention, dense and routed-expert writers on their own removes 0.039, 0.016 and 0.148 of refusal respectively; editing all three together removes 0.776. As a result, 74% of the effect exists only under the joint intervention. The part the conventional recipe reaches by module-name matching accounts for 0.066 of that 0.776, which is why it fails silently on an MoE. The effect does not follow from removing just any direction: ablating a random direction orthogonal to it leaves refusal unchanged. A category-concentrated residue survives every edit we tried: subspaces fitted on violence, sexual content and hate leave measurable refusal at every rank from 1 to 12. We report the method, the 41-89 percentage-point reductions it achieves across seven harmful benchmarks with no detected change in capability, and the boundary where it stops.

How do the methods used to train language models to refuse harmful requests shape how that refusal actually works inside the model? We compare three post-training methods - supervised fine-tuning, reasoning-augmented fine-tuning (training on reasoning chains that justify a safety decision), and preference optimization (ORPO) - across three architecturally distinct models (Llama-3.1-8B, Gemma-2-9B, Qwen3-8B). We find that training method, not just data, reshapes how refusal is computed internally: reasoning-augmented training consistently produces a distinct kind of refusal computation, visible across all three models, while architecture independently shapes internal structure and how reliably refusal can be steered. Most importantly, no method we study achieves all three properties we would want from safe alignment at once: refusal that isn't concentrated in a few fragile components, safety gains that don't cost general capability, and safety behavior correctable through small, targeted edits. We caution against treating current post-training methods as a solved, reliable defense, especially for security-critical use. Code and models are available in https://github.com/hoangcuongnguyen2001/Beyond-Shallow-Alignment.

We propose a criterion for discrimination against a specified sensitive attribute in supervised learning, where the goal is to predict some target based on available features. Assuming data about the predictor, target, and membership in the protected group are available, we show how to optimally adjust any learned predictor so as to remove discrimination according to our definition. Our framework also improves incentives by shifting the cost of poor classification from disadvantaged groups to the decision maker, who can respond by improving the classification accuracy. In line with other studies, our notion is oblivious: it depends only on the joint statistics of the predictor, the target and the protected attribute, but not on interpretation of individualfeatures. We study the inherent limits of defining and identifying biases based on such oblivious measures, outlining what can and cannot be inferred from different oblivious tests. We illustrate our notion using a case study of FICO credit scores.

Large Language Models (LLMs) require careful safety alignment to prevent malicious outputs. While significant research focuses on mitigating harmful content generation, the enhanced safety often come with the side effect of over-refusal, where LLMs may reject innocuous prompts and become less helpful. Although the issue of over-refusal has been empirically observed, a systematic measurement is challenging due to the difficulty of crafting prompts that appear harmful but are benign. This study proposes a novel method for automatically generating large-scale sets of "seemingly toxic prompts" (benign prompts likely rejected by LLMs). Leveraging this technique, we introduce OR-Bench, the first large-scale over-refusal benchmark. OR-Bench comprises 80,000 seemingly toxic prompts across 10 common rejection categories, a subset of around 1,000 hard prompts that are challenging even for state-of-the-art LLMs, and an additional 600 toxic prompts to prevent indiscriminate responses. We then conduct a comprehensive study to measure the over-refusal of 25 popular LLMs across 8 model families. Our datasets are available at https://huggingface.co/datasets/bench-llm/or-bench and the demo can be found at https://huggingface.co/spaces/bench-llm/or-bench. We hope this benchmark can help the community develop better safety aligned models.

Inasmuch as the removal of refusal behavior from instruction-tuned language models by directional abliteration requires the extraction of refusal-mediating directions from the residual stream activation space, and inasmuch as the construction of the contrast baseline against which harmful prompt activations are compared has been treated in the existing literature as an implementation detail rather than a methodological concern, the present work investigates whether a topically matched contrast baseline yields superior refusal directions. The investigation is carried out on the Qwen~3.5 2B model using per-category matched prompt pairs, per-class Self-Organizing Map extraction, and Singular Value Decomposition orthogonalization. It was found that topic-matched contrast produces no functional refusal directions at any tested weight level on any tested layer, while unmatched contrast on the same model, same extraction code, and same evaluation protocol achieves complete refusal elimination on six layers. The geometric analysis of the failure establishes that topic-matched subtraction cancels the dominant activation component shared between harmful and harmless prompts of the same subject, reducing the extracted direction magnitude below the threshold at which weight-matrix projection perturbs the residual stream. The implications for the design of contrast baselines in abliteration research are discussed.

Safety guardrails in open-weight language models can be readily bypassed using Refusal Feature Ablation (RFA), a technique that identifies and projects out a linear refusal direction from the residual stream, often achieving a high attack success rate (ASR) while preserving model capability. Defending against these attacks typically requires computationally expensive safety finetuning for every new checkpoint. We introduce Decoy Direction Optimization (DDO), a fast, post-hoc weight-editing defense that requires no base-model finetuning. Our approach is based on a simple mechanistic insight: ablation attacks rely on contrastive estimators to find the refusal direction. Rather than trying to hide the true refusal circuitry, DDO actively injects a high-magnitude, nonlinear decoy signal into the network's MLP neurons. When an attacker attempts to locate the refusal direction, the decoy corrupts their estimator, tricking them into ablating a harmless orthogonal feature while the actual safety mechanism remains intact. We prove a spectral bound formalizing this effect and evaluate DDO across six model families, achieving <10% ASR under standard RFA. On Llama-3-8B-Instruct, DDO remains comparable to trained defenses under adaptive multi-phase attacks (65% vs. 58% worst-case ASR) and reduces Heretic weight-level attack ASR from 88.7% to 18%, all at 30 to 450 times lower optimization cost per configuration than the trained baselines.

Current alignment evaluation mostly measures whether models encode dangerous concepts and whether they refuse harmful requests. Both miss the layer where alignment often operates: routing from concept detection to behavioral policy. We study political censorship in Chinese-origin language models as a natural experiment, using probes, surgical ablations, and behavioral tests across nine open-weight models from five labs. Three findings follow. First, probe accuracy alone is non-diagnostic: political probes, null controls, and permutation baselines can all reach 100%, so held-out category generalization is the informative test. Second, surgical ablation reveals lab-specific routing. Removing the political-sensitivity direction eliminates censorship and restores accurate factual output in most models tested, while one model confabulates because its architecture entangles factual knowledge with the censorship mechanism. Cross-model transfer fails, indicating that routing geometry is model- and lab-specific. Third, refusal is no longer the dominant censorship mechanism. Within one model family, hard refusal falls to zero while narrative steering rises to the maximum, making censorship invisible to refusal-only benchmarks. These results support a three-stage descriptive framework: detect, route, generate. Models often retain the relevant knowledge; alignment changes how that knowledge is expressed. Evaluations that audit only detection or refusal therefore miss the routing mechanism that most directly determines behavior.

Following the recent achievement of gold-medal performance on the IMO by frontier LLMs, the community is searching for the next meaningful and challenging target for measuring LLM reasoning. Whereas olympiad-style problems measure step-by-step reasoning alone, research-level problems use such reasoning to advance the frontier of mathematical knowledge itself, emerging as a compelling alternative. Yet research-level math benchmarks remain scarce because such problems are difficult to source (e.g., Riemann Bench and FrontierMath-Tier 4 contain 25 and 50 problems, respectively). To support reliable evaluation of next-generation frontier models, we introduce Soohak, a 439-problem benchmark newly authored from scratch by 64 mathematicians. Soohak comprises two subsets. On the Challenge subset, frontier models including Gemini-3-Pro, GPT-5, and Claude-Opus-4.5 reach 30.4%, 26.4%, and 10.4% respectively, leaving substantial headroom, while leading open-weight models such as Qwen3-235B, GPT-OSS-120B, and Kimi-2.5 remain below 15%. Notably, beyond standard problem solving, Soohak introduces a refusal subset that probes a capability intrinsic to research mathematics: recognizing ill-posed problems and pausing rather than producing confident but unjustified answers. On this subset, no model exceeds 50%, identifying refusal as a new optimization target that current models do not directly address. To prevent contamination, the dataset will be publicly released in late 2026, with model evaluations available upon request in the interim.

Frontier large language models are increasingly deployed as orchestration backbones for biological research workflows, yet no shared evidence base exists for comparing their refusal behaviour on legitimate research prompts. RefusalBench, introduced here, is a matched-triple benchmark of 141 prompts in 47 bundles that holds task framing constant while varying only biological risk tier (benign, borderline, dual-use), enabling tier-conditioned comparisons robust to subdomain confounding. A 15-prompt should-refuse positive-control module establishes per-model calibration floors; three models fail to refuse even these prompts. Across 19 frontier models in the May 2026 snapshot, strict refusal rates span 0.1% to 94.6% on identical prompts. Jurisdiction does not predict refusal in this snapshot (Mann-Whitney U, p = 0.393; EU n = 1, US bimodal); provider identity does, with Anthropic's API stack predicting refusal at OR = 21.03 (95% CI: 14.58-30.34 prompt-clustered; 5.70-77.55 under model-clustered GEE). This effect is best read as access-path-level rather than model-weight-level: 99.8% of Anthropic's strict refusals carry the same safety_policy adjudicated reason code, consistent with a small set of canonical refusal templates rather than case-by-case model reasoning. Strict refusal rate misranks safety calibration: Grok 4.20 achieves the highest tier discrimination (Youden's J = 0.787) while ranking only seventh by overall refusal rate, and Claude Opus 4.7's J dropped 65% from prior versions with no improvement in dual-use detection. Nine of 18 frontier models exhibit a hedge-but-help partial-compliance pattern at dual-use tier that binary refusal metrics cannot detect.

Large language models typically undergo post-training to align them with safety policies but there exist many sophisticated jailbreaks that sidestep established safeguards. For instance, prior work by Andriushchenko et al. (2025) has found that changing the grammatical tense from present to past can be enough to elicit harmful responses. In this work, we uncover a more general failure of non-imperative syntactic forms. We demonstrate that this syntactic vulnerability exists in 16 models up to 70B parameters, using behavioral evaluation. To investigate the root cause, we apply causal mediation analysis, finding that refusal is partially conditioned on upstream syntactic features. By steering these purely syntactic features we are able to trigger and suppress refusal. Finally, we trace this ill-conditioning to linguistically biased post-training data of open-source models and show that increasing syntactic diversity can mitigate the issue. Our findings suggest that current alignment approaches introduce confounders that prevent a pure semantic grounding of the refusal decision.

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.

Arditi et al. (2024) has shown that refusal in safety fine-tuned chat models is mediated by a single linear direction in the residual stream, recoverable by a difference-in-means (DiM) of harmful and harmless activations. We compare DiM-based interventions (activation addition and directional ablation) with two interventions derived from Iterative Nullspace Projection (INLP) -- nullspace projection and counterfactual flipping -- on five open-weight chat models, asking whether INLP can match DiM at steering refusal and whether its richer parameterisation yields more tweakable interventions. INLP counterfactual flipping is competitive with DiM directional ablation on refusal suppression, while nullspace projection is consistently weaker. Restricting INLP to the leading directions of the extracted subspace preserves most of the suppression effect at near-baseline perplexity, giving a tunable capability. Geometrically, the two INLP interventions land in qualitatively different regions of activation space: nullspace projection collapses transformed activations between the harmful and harmless clusters, while counterfactual flipping moves them into the opposite cluster, suggesting that the model encodes the absence of a concept differently from its opposite -- an intriguing distinction that warrants further investigation in future work.

Large language models (LLMs) frequently produce false refusals, declining benign requests that contain terms resembling unsafe queries. We address this challenge by introducing two comprehensive benchmarks: the Exaggerated Safety Benchmark (XSB) for single-turn prompts, annotated with "Focus" keywords that identify refusal-inducing triggers, and the Multi-turn Scenario-based Exaggerated Safety Benchmark (MS-XSB), which systematically evaluates refusal calibration in realistic, context-rich dialog settings. Our benchmarks reveal that exaggerated refusals persist across diverse recent LLMs and are especially pronounced in complex, multi-turn scenarios. To mitigate these failures, we leverage post-hoc explanation methods to identify refusal triggers and deploy three lightweight, model-agnostic approaches, ignore-word instructions, prompt rephrasing, and attention steering, at inference time, all without retraining or parameter access. Experiments on four instruction-tuned Llama models demonstrate that these strategies substantially improve compliance on safe prompts while maintaining robust safety protections. Our findings establish a reproducible framework for diagnosing and mitigating exaggerated refusals, highlighting practical pathways to safer and more helpful LLM deployments.

Refusal rates are a poor proxy for LLM safety, i.e., a model may over-refuse benign prompts while still complying with harmful ones. We audit both failure modes across 21 open-weight LLMs on four safety benchmarks (OR-Bench, XSTest, ToxiGen, BOLD), using a composition adjustment to isolate model sensitivity from dataset toxicity confounds. We report three findings. First, models adopt fundamentally different calibration strategies: conservative ecosystems such as Llama suppress unsafe outputs at the cost of elevated over-refusals, while permissive ecosystems such as DeepSeek and Qwen preserve helpfulness but tolerate higher harmful compliance. Second, demographic protection is unequal: models over-protect prominent racial and religious groups, frequently refusing even benign prompts about them, while providing substantially weaker protection against disability-targeted attacks. Third, refusal and compliance tendencies are stable within model families across generations and scales, suggesting that post-training objectives shape safety behavior more than architecture. Our results call for joint, demographically-aware, and multi-judge safety evaluation.

Diffusion language models (DLMs) have recently emerged as a promising alternative to autoregressive (AR) models, offering parallel decoding and controllable sampling dynamics while achieving competitive generation quality at scale. Despite this progress, the role of sampling mechanisms in shaping refusal behavior and jailbreak robustness remains poorly understood. In this work, we present a fundamental analytical framework for step-wise refusal dynamics, enabling comparison between AR and diffusion sampling. Our analysis reveals that the sampling strategy itself plays a central role in safety behavior, as a factor distinct from the underlying learned representations. Motivated by this analysis, we introduce the Step-Wise Refusal Internal Dynamics (SRI) signal, which supports interpretability and improved safety for both AR and DLMs. We demonstrate that the geometric structure of SRI captures internal recovery dynamics, and identifies anomalous behavior in harmful generations as cases of incomplete internal recovery that are not observable at the text level. This structure enables lightweight inference-time detectors that generalize to unseen attacks while matching or outperforming existing defenses with over 100times lower inference overhead.

Refusal-Aware Instruction Tuning (RAIT) aims to enhance Large Language Models (LLMs) by improving their ability to refuse responses to questions beyond their knowledge, thereby reducing hallucinations and improving reliability. Effective RAIT must address two key challenges: firstly, effectively reject unknown questions to minimize hallucinations; secondly, avoid over-refusal to ensure questions that can be correctly answered are not rejected, thereby maintain the helpfulness of LLM outputs. In this paper, we address the two challenges by deriving insightful observations from the gradient-based perspective, and proposing the Gradient-driven Refusal Aware Instruction Tuning Framework GRAIT: (1) employs gradient-driven sample selection to effectively minimize hallucinations and (2) introduces an adaptive weighting mechanism during fine-tuning to reduce the risk of over-refusal, achieving the balance between accurate refusals and maintaining useful responses. Experimental evaluations on open-ended and multiple-choice question answering tasks demonstrate that GRAIT significantly outperforms existing RAIT methods in the overall performance. The source code and data will be available at https://github.com/opendatalab/GRAIT .

Safety alignment in large language models relies predominantly on English-language training data. When harmful intent is expressed in low-resource languages, refusal mechanisms that hold in English frequently fail to activate. We introduce LSR (Linguistic Safety Robustness), the first systematic benchmark for measuring cross-lingual refusal degradation in West African languages: Yoruba, Hausa, Igbo, and Igala. LSR uses a dual-probe evaluation protocol - submitting matched English and target-language probes to the same model - and introduces Refusal Centroid Drift (RCD), a metric that quantifies how much of a model's English refusal behavior is lost when harmful intent is encoded in a target language. We evaluate Gemini 2.5 Flash across 14 culturally grounded attack probes in four harm categories. English refusal rates hold at approximately 90 percent. Across West African languages, refusal rates fall to 35-55 percent, with Igala showing the most severe degradation (RCD = 0.55). LSR is implemented in the Inspect AI evaluation framework and is available as a PR-ready contribution to the UK AISI's inspect_evals repository. A live reference implementation and the benchmark dataset are publicly available.

Omni-modal Large Language Models (OLLMs) greatly expand LLMs' multimodal capabilities but also introduce cross-modal safety risks. However, a systematic understanding of vulnerabilities in omni-modal interactions remains lacking. To bridge this gap, we establish a modality-semantics decoupling principle and construct the AdvBench-Omni dataset, which reveals a significant vulnerability in OLLMs. Mechanistic analysis uncovers a Mid-layer Dissolution phenomenon driven by refusal vector magnitude shrinkage, alongside the existence of a modal-invariant pure refusal direction. Inspired by these insights, we extract a golden refusal vector using Singular Value Decomposition and propose OmniSteer, which utilizes lightweight adapters to modulate intervention intensity adaptively. Extensive experiments show that our method not only increases the Refusal Success Rate against harmful inputs from 69.9% to 91.2%, but also effectively preserves the general capabilities across all modalities. Our code is available at: https://github.com/zhrli324/omni-safety-research.

Text-to-image (T2I) models have significantly advanced in producing high-quality images. However, such models have the ability to generate images containing not-safe-for-work (NSFW) content, such as pornography, violence, political content, and discrimination. To mitigate the risk of generating NSFW content, refusal mechanisms, i.e., safety checkers, have been developed to check potential NSFW content. Adversarial prompting techniques have been developed to evaluate the robustness of the refusal mechanisms. The key challenge remains to subtly modify the prompt in a way that preserves its sensitive nature while bypassing the refusal mechanisms. In this paper, we introduce TokenProber, a method designed for sensitivity-aware differential testing, aimed at evaluating the robustness of the refusal mechanisms in T2I models by generating adversarial prompts. Our approach is based on the key observation that adversarial prompts often succeed by exploiting discrepancies in how T2I models and safety checkers interpret sensitive content. Thus, we conduct a fine-grained analysis of the impact of specific words within prompts, distinguishing between dirty words that are essential for NSFW content generation and discrepant words that highlight the different sensitivity assessments between T2I models and safety checkers. Through the sensitivity-aware mutation, TokenProber generates adversarial prompts, striking a balance between maintaining NSFW content generation and evading detection. Our evaluation of TokenProber against 5 safety checkers on 3 popular T2I models, using 324 NSFW prompts, demonstrates its superior effectiveness in bypassing safety filters compared to existing methods (e.g., 54%+ increase on average), highlighting TokenProber's ability to uncover robustness issues in the existing refusal mechanisms.

Autonomous agents increasingly perform bounded software tasks under an orchestrator that retries, resumes, and budgets them. The machinery such orchestrators reach for is the service mesh's: retry, timeout, and error-rate circuit breaking. We report a failure study of a production agentic software-delivery platform over 147 numbered incidents spanning 81 runs, each with a measured cost and, in most cases, a mutation proof reproducing the failure. All three assumptions those primitives rest on are violated in practice, and we quantify the consequences: a loop of fifty-four consecutive successful tool calls no error-rate breaker could see; a progress signal constant by construction, guaranteeing a false trip on the third repair round and driving one run from six of six components to three; twenty-one events accumulated across six invocations of one delegation, making a correct, idempotent component unwinnable; a misrouted failure that woke five components for a two-component fault, leaving three bystanders regressing working code; and twelve incidents in which the enforcement layer blocked correct work, the most expensive costing 107 agent turns and zero accepted writes. We find one cross-cutting cause and its dual. Identity adequacy: in five separate subsystems an identity that failed to discriminate produced a confident wrong answer, and two of them derived the corrective rule independently. Evidence adequacy: a reliability decision may be taken only on evidence capable of moving, attributable to what it measures, and deterministic under identical conditions. From the findings we derive seven reliability primitives whose enforcement unit is the delegation rather than the message, and specify the controlled evaluation the study motivates but does not constitute.

Responsible practices for deploying language models include guiding models to recognize and refuse answering prompts that are considered unsafe, while complying with safe prompts. Achieving such behavior typically requires updating model weights, which is costly and inflexible. We explore opportunities to steering model activations at inference time, which does not require updating weights. Using sparse autoencoders, we identify and steer features in Phi-3 Mini that mediate refusal behavior. We find that feature steering can improve Phi-3 Minis robustness to jailbreak attempts across various harms, including challenging multi-turn attacks. However, we discover that feature steering can adversely affect overall performance on benchmarks. These results suggest that identifying steerable mechanisms for refusal via sparse autoencoders is a promising approach for enhancing language model safety, but that more research is needed to mitigate feature steerings adverse effects on performance.

Block drafters propose several tokens in one forward pass, before earlier target tokens are realised. Their rejection mixes two losses: missing within-block path information and imperfect modelling of observable information. Accepted length cannot distinguish them. We separate the two with an information floor, the minimum expected rejection at a specified conditioning order; rejection above this floor is the model gap. Estimating both from target rollouts across four domains, four open-weight targets, and a frontier API target yields three findings. First, the all-parallel floor reaches 0.286 at the final slot on Qwen3-4B, limiting even the best proposal to 71% per-slot acceptance. Second, one realised token removes 86--100% of this floor, a locality also recovered by an independent mutual-information analysis. Third, current drafters remain far above their floors: the final-slot model gap accounts for 43--64% of DFlash rejection and 85--92% of DSpark's oracle-conditioned rejection. These findings separate the value of short-range conditioning from proposal quality. Guided by this distinction, we replace a context-independent predecessor correction with a prefix-attention head that reads committed context conditional on the predecessor. With supporting backbone components, the resulting Qwen3-4B drafter improves mean serving accepted length by 3.83% over the released DSpark checkpoint across nine tasks, without increasing the conditioning order.

Alignment tuning is meant to make harmful-request refusal robust, yet this safety behavior can be erased by a small set of benign fine-tuning examples. This is a deployment risk for open-weight models because a checkpoint can pass refusal tests at release time and later lose refusal under low-cost downstream fine-tuning. Prior work has established these refusal failures, but existing studies do not show how to detect this fragility in the aligned model itself before an attack or fine-tuning intervention is run. We introduce Skin-Deep, a geometric diagnostic that detects alignment fragility directly from the aligned model's hidden-state activations before such an intervention is run and compresses the layer-wise safety geometry into a single scalar, the Geometric Fragility Score (GFS). Applied to twenty-one instruction-tuned models spanning six alignment recipes and 3B--32B parameters, Skin-Deep reveals a recurring low-rank safety subspace across model families. Direction ablations show that removing directions in this subspace weakens harmful-request refusal, providing causal evidence that the recovered geometry underlies refusal behavior. Crucially, GFS identifies, before any fine-tuning, the initially safe model that retains the most refusal after small-scale LoRA fine-tuning. These results establish GFS as a practical pre-deployment diagnostic for flagging fragile refusal behavior without running an attack.

Leading language model (LM) providers like OpenAI and Google offer fine-tuning APIs that allow customers to adapt LMs for specific use cases. To prevent misuse, these LM providers implement filtering mechanisms to block harmful fine-tuning data. Consequently, adversaries seeking to produce unsafe LMs via these APIs must craft adversarial training data that are not identifiably harmful. We make three contributions in this context: 1. We show that many existing attacks that use harmless data to create unsafe LMs rely on eliminating model refusals in the first few tokens of their responses. 2. We show that such prior attacks can be blocked by a simple defense that pre-fills the first few tokens from an aligned model before letting the fine-tuned model fill in the rest. 3. We describe a new data-poisoning attack, ``No, Of course I Can Execute'' (NOICE), which exploits an LM's formulaic refusal mechanism to elicit harmful responses. By training an LM to refuse benign requests on the basis of safety before fulfilling those requests regardless, we are able to jailbreak several open-source models and a closed-source model (GPT-4o). We show an attack success rate (ASR) of 57% against GPT-4o; our attack earned a Bug Bounty from OpenAI. Against open-source models protected by simple defenses, we improve ASRs by an average of 3.25 times compared to the best performing previous attacks that use only harmless data. NOICE demonstrates the exploitability of repetitive refusal mechanisms and broadens understanding of the threats closed-source models face from harmless data.

Safety-aligned large language models are trained to refuse harmful requests, yet embedding the same requests in particular scenarios can bypass their safeguards. Existing red-teaming methods empirically identify effective scenarios through observed attack outcomes, but why particular scenarios weaken refusal remains mechanistically unclear. Meanwhile, mechanistic interpretability studies have characterized both refusal directions and jailbreak-associated features, without explaining the relationship between the two representations. In this work, we show that scenario-wrapped prompts activate internal scenario directions whose causal steering consistently reduces refusal scores. Building on this finding, we propose Concept2Scenario, a concept-based attribution framework for vulnerable scenario discovery. It instantiates a broad concept space with a sparse autoencoder, attributes refusal suppression to individual concepts, translates the identified concepts into interpretable natural-language scenarios, and identifies synergistic scenario combinations through interaction attribution. Across three open-source models, two safety benchmarks, and six black-box jailbreak methods, the discovered scenarios serve as reusable priors that improve average attack success rates by up to 18.2 percentage points. They also transfer to GPT-5, Claude-Haiku-4.5, and Gemini-3-Flash, suggesting that some scenario-level refusal vulnerabilities are shared across model families. Moreover, the identified combinations outperform their individual constituents and enable iterative attacks to succeed in fewer turns.

A central aspiration of mechanistic interpretability is controllability: if we know where a behavior is represented in a model's activations, we should be able to modify it. This rests on a hidden premise -- that the direction which detects a behavior and the direction which controls it are the same, or close. We test this geometrically: what is the angle between the direction that best detects a behavior and the one that best causes it? If detection implies control the cosine is near 1; otherwise it quantifies a detection-intervention gap. On Gemma 2-2B-it, output format (clean JSON vs markdown fencing) collapses both roles onto one axis. Hallucination does not: the model detects fake entities with perfect linear separability (AUC = 1.000 from layer 5), yet that direction sits at cos = 0.12 (about 83 degrees) from the direction producing a refusal -- a small, reproducible alignment, far from the cos = 1 that "detection is control" would require. A detector built from activations, with no chosen tokens, likewise fails to align (cos = -0.06). The gap generalizes: across four models from three families and two scales (1B-9B), cos stays in [0.12, 0.20], identical before and after instruction tuning (0.1197 vs 0.1200), placing its origin in pretraining. A 15-degree rotation toward the refusal direction partially bridges it -- 73% and 60% refusal on two held-out fake-entity categories at 1.8% false positives. We then ask whether this cosine predicts steerability, and it does not: detection is a high-dimensional class, not a single direction, and what separates the steerable case is functional, not readable from a static angle. The cosine is a weight-computable signature of the dissociation between knowing and steering, not a predictor of it.

While activation steering in large language models (LLMs) is a growing area of research, methods can often incur broader effects than desired. This motivates isolation of purer concept vectors to enable targeted interventions and understand LLM behavior at a more granular level. We present RepIt, a simple and data-efficient framework for isolating concept-specific representations. Across five frontier LLMs, RepIt enables precise interventions: it selectively suppresses refusal on targeted concepts while preserving refusal elsewhere, producing models that answer WMD-related questions while still scoring as safe on standard benchmarks. We further show that the corrective signal localizes to just 100-200 neurons and that robust target representations can be extracted from as few as a dozen examples on a single A6000. This efficiency raises a dual concern: manipulations can be performed with modest compute and data to extend to underrepresented data-scarce topics while evading existing benchmarks. By disentangling refusal vectors with RepIt, this work demonstrates that targeted interventions can counteract overgeneralization, laying the foundation for more granular control of model behavior.

Existing training-time safety alignment techniques for large language models (LLMs) remain vulnerable to jailbreak attacks. Direct preference optimization (DPO), a widely deployed alignment method, exhibits limitations in both experimental and theoretical contexts as its loss function proves suboptimal for refusal learning. Through gradient-based analysis, we identify these shortcomings and propose an improved safety alignment that disentangles DPO objectives into two components: (1) robust refusal training, which encourages refusal even when partial unsafe generations are produced, and (2) targeted unlearning of harmful knowledge. This approach significantly increases LLM robustness against a wide range of jailbreak attacks, including prefilling, suffix, and multi-turn attacks across both in-distribution and out-of-distribution scenarios. Furthermore, we introduce a method to emphasize critical refusal tokens by incorporating a reward-based token-level weighting mechanism for refusal learning, which further improves the robustness against adversarial exploits. Our research also suggests that robustness to jailbreak attacks is correlated with token distribution shifts in the training process and internal representations of refusal and harmful tokens, offering valuable directions for future research in LLM safety alignment. The code is available at https://github.com/wicai24/DOOR-Alignment

Large language models (LLMs) have shown remarkable performance in various tasks but often fail to handle queries that exceed their knowledge and capabilities, leading to incorrect or fabricated responses. This paper addresses the need for LLMs to recognize and refuse infeasible tasks due to the required skills surpassing their capabilities. We first systematically conceptualize infeasible tasks for LLMs, providing formal definitions and categorizations that cover a spectrum of related hallucinations. We develop and benchmark a new dataset comprising diverse infeasible and feasible tasks to test multiple LLMs' abilities on task feasibility. Furthermore, we explore the potential of training enhancements to increase LLMs' refusal capabilities with fine-tuning. Experiments validate the effectiveness of our methods, offering promising directions for refining the operational boundaries of LLMs in real applications.

Can a language model estimate its familiarity with an entity before generating an answer? We study activations at the final prompt token in twelve instruction-tuned models from the Bielik, PLLuM, Gemma-4, and Qwen3 families, using a new dataset of 1,440 Polish entities spanning four domains and ten Wikipedia-pageview deciles, plus fabricated controls. Familiarity-probe scores separate real from fabricated entities in every family; in the Polish-adapted Bielik and PLLuM families they additionally track entity popularity (model-mean Spearman ρ 0.28-0.57, versus at most 0.11 in Gemma-4 and Qwen3), a pattern more strongly associated with Polish adaptation than with parameter count in this model sample. In a paired experiment on two families, probes retain 96-101% of within-language AUROC when the Polish question stem is replaced with an English one around unchanged entity names, showing robustness to prompt language in this setting. In Gemma-4-12B, the only model that natively refuses, adding a one-dimensional familiarity direction at a single layer moves refusal rates monotonically in both directions (0.24 to 1.00 on well-known entities; 0.73 to 0.00 on unknown ones). Finally, a calibrated familiarity probe is competitive among pre-generation abstention gates, although post-generation detectors better predict behavioral error on average. These results support a graded pre-generation entity-familiarity readout, and a separation between representational familiarity and the policy that converts it into abstention.

This study addresses a critical gap in safety tuning practices for Large Language Models (LLMs) by identifying and tackling a refusal position bias within safety tuning data, which compromises the models' ability to appropriately refuse generating unsafe content. We introduce a novel approach, Decoupled Refusal Training (DeRTa), designed to empower LLMs to refuse compliance to harmful prompts at any response position, significantly enhancing their safety capabilities. DeRTa incorporates two novel components: (1) Maximum Likelihood Estimation (MLE) with Harmful Response Prefix, which trains models to recognize and avoid unsafe content by appending a segment of harmful response to the beginning of a safe response, and (2) Reinforced Transition Optimization (RTO), which equips models with the ability to transition from potential harm to safety refusal consistently throughout the harmful response sequence. Our empirical evaluation, conducted using LLaMA3 and Mistral model families across six attack scenarios, demonstrates that our method not only improves model safety without compromising performance but also surpasses well-known models such as GPT-4 in defending against attacks. Importantly, our approach successfully defends recent advanced attack methods (e.g., CodeAttack) that have jailbroken GPT-4 and LLaMA3-70B-Instruct. Our code and data can be found at https://github.com/RobustNLP/DeRTa.

While Large Language Models (LLMs) have achieved remarkable performance, they remain vulnerable to jailbreak attacks that circumvent safety constraints. Existing strategies, ranging from heuristic prompt engineering to computationally intensive optimization, often face significant trade-offs between effectiveness and efficiency. In this work, we propose Contextual Representation Ablation (CRA), a novel inference-time intervention framework designed to dynamically silence model guardrails. Predicated on the geometric insight that refusal behaviors are mediated by specific low-rank subspaces within the model's hidden states, CRA identifies and suppresses these refusal-inducing activation patterns during decoding without requiring expensive parameter updates or training. Empirical evaluation across multiple safety-aligned open-source LLMs demonstrates that CRA significantly outperforms baselines. These results expose the intrinsic fragility of current alignment mechanisms, revealing that safety constraints can be surgically ablated from internal representations, and underscore the urgent need for more robust defenses that secure the model's latent space.

Large reasoning models (LRMs) with multi-step reasoning capabilities have shown remarkable problem-solving abilities, yet they exhibit concerning safety vulnerabilities that remain poorly understood. In this work, we investigate why safety alignment fails in reasoning models through a mechanistic interpretability lens. Using a linear probing approach to trace refusal intentions across token positions, we discover a striking phenomenon termed as refusal cliff: many poorly-aligned reasoning models correctly identify harmful prompts and maintain strong refusal intentions during their thinking process, but experience a sharp drop in refusal scores at the final tokens before output generation. This suggests that these models are not inherently unsafe; rather, their refusal intentions are systematically suppressed. Through causal intervention analysis, we identify a sparse set of attention heads that negatively contribute to refusal behavior. Ablating just 3\% of these heads can reduce attack success rates below 10\%. Building on these mechanistic insights, we propose Cliff-as-a-Judge, a novel data selection method that identifies training examples exhibiting the largest refusal cliff to efficiently repair reasoning models' safety alignment. This approach achieves comparable safety improvements using only 1.7\% of the vanilla safety training data, demonstrating a less-is-more effect in safety alignment.

We consider the problem of selective prediction (also known as reject option) in deep neural networks, and introduce SelectiveNet, a deep neural architecture with an integrated reject option. Existing rejection mechanisms are based mostly on a threshold over the prediction confidence of a pre-trained network. In contrast, SelectiveNet is trained to optimize both classification (or regression) and rejection simultaneously, end-to-end. The result is a deep neural network that is optimized over the covered domain. In our experiments, we show a consistently improved risk-coverage trade-off over several well-known classification and regression datasets, thus reaching new state-of-the-art results for deep selective classification.

We study inference-time pattern-memory gating in a production-scale clinical natural language processing (NLP) pipeline. The pipeline pairs a generator (Llama-3.3 70B) proposing extractions with a verifier (MMed-Llama-3.1 70B) accepting or rejecting them, over 167,034 PMC-Patients narratives, and adds a lightweight memory that learns at deployment which extractions to filter, so the verifier need not re-examine candidates already seen to fail. We report four findings. First, learning filtering rules directly from the verifier's rejections failed at full scale: the relation-extraction filter stayed empty despite 785,797 logged rejections, because they were spread too thinly across too many distinct forms to accumulate. Second, a simpler rule using a fixed clinical ontology produced the same filtering without the verifier, capturing 49,734 ontology-violating relations on a held-out 5,000-patient set. Third, of five versions of the question-answering filter, four failed for distinct, instructive reasons; the fifth succeeded by checking whether a patient's extracted entities support the question asked, and where it applies was 1.84 times likelier to flag an answer the verifier would reject than one it would accept. Fourth, one pattern held across all five: a filter is selective only when it tests the same evidence the verifier weighs, not when it imitates the verifier's output. Together these give a transferable result for any generator-verifier pipeline: the most natural memory design can fail silently at scale, and whether a pre-generation gate is selective is decided before any engineering effort, by whether its signal probes the question the verifier itself answers. Throughout, the system flags suspect extractions rather than deleting them, so every decision stays visible for clinical review. All code and test artefacts are released openly.

Large language model safety evaluation remains heavily English-centered, leaving low-resource languages under-measured even when models are deployed globally. We evaluate four open-weight instruction-tuned models on SomaliBench v0, a native-author-verified benchmark of 100 harmful-intent prompts paired across English and Somali. Each of Llama-3.1-8B-Instruct, Gemma-2-9B-Instruct, Qwen-2.5-7B-Instruct, and Aya-23-8B is run locally with temperature 0 and the same English "helpful, harmless, and honest" (HHH) system prompt. A pinned Claude Sonnet snapshot (claude-sonnet-4-5-20250929) classifies each response as refused, complied, or unclear; the native author spot-checks a stratified 80-row sample. We find large English-to-Somali refusal gaps for all four models: Llama-3.1-8B (0.90; 95% bootstrap CI [0.85, 0.96]), Aya-23-8B (0.75 [0.67, 0.83]), Qwen-2.5-7B (0.69 [0.59, 0.78]), and Gemma-2-9B (0.38 [0.27, 0.49]). For three models, the dominant Somali non-refusal mode is not fluent harmful compliance but unclear output: empty, wrong-language, or incoherent generations. The native verification spot-check achieves 100% agreement with the judge (Cohen's kappa = 1.00) on the 80 sampled rows. We report aggregate refusal rates, category gaps, and reliability statistics only; raw model generations are retained locally and are not released.

In critical applications, it is vital for classifiers to defer decision-making to humans. We propose a post-hoc method that makes existing classifiers selectively abstain from predicting certain samples. Our abstaining classifier is incentivized to maintain the original accuracy for each sub-population (i.e. no harm) while achieving a set of group fairness definitions to a user specified degree. To this end, we design an Integer Programming (IP) procedure that assigns abstention decisions for each training sample to satisfy a set of constraints. To generalize the abstaining decisions to test samples, we then train a surrogate model to learn the abstaining decisions based on the IP solutions in an end-to-end manner. We analyze the feasibility of the IP procedure to determine the possible abstention rate for different levels of unfairness tolerance and accuracy constraint for achieving no harm. To the best of our knowledge, this work is the first to identify the theoretical relationships between the constraint parameters and the required abstention rate. Our theoretical results are important since a high abstention rate is often infeasible in practice due to a lack of human resources. Our framework outperforms existing methods in terms of fairness disparity without sacrificing accuracy at similar abstention rates.

Perturbation probing generates task-specific causal hypotheses for FFN neurons in large language models using two forward passes per prompt and no backpropagation, followed by a one-time intervention sweep of about 150 passes amortized across all identified neurons. Across eight behavioral circuits, 13 models, and four architecture families, we identify two circuit structures that organize LLM behavior. Opposition circuits appear when RLHF suppresses a pre-training tendency. In safety refusal, about 50 neurons, or 0.014 percent of all neurons, control the refusal template; ablating them changes 80 percent of response formats on 520 AdvBench prompts while producing near-zero harmful compliance, 3 of 520 cases, all with disclaimers. Routing circuits appear for pre-training behaviors distributed through attention. For language selection, residual-stream direction injection switches English to Chinese output on 99.1 percent of 580 benchmark prompts in the 3 of 19 tested models that satisfy three observed conditions: bilingual training, FFN-to-skip signal ratio between 0.3 and 1.1, and linear representability. The same intervention fails on the other 16 models and on math, code, and factual circuits, defining the limits of directional steering. The FFN-to-skip signal ratio, computed from the same two forward passes, distinguishes the two structures and predicts the appropriate intervention. Circuit topology varies by architecture, from Qwen's concentrated FFN bottleneck to Gemma's normalization-shielded circuit. In Qwen3.5-2B, ablating 20 neurons eliminates multi-turn sycophantic capitulation, while amplifying 10 related neurons improves factual correction from 52 percent to 88 percent on 200 TruthfulQA prompts. These results show that perturbation probing offers mechanistic insight into RLHF-organized behavior and a practical toolkit for precision template-layer editing.

Safety-aligned language models refuse harmful requests through learned refusal behaviors encoded in their internal representations. Recent activation-based jailbreaking methods circumvent these safety mechanisms by applying orthogonal projections to remove refusal directions, but these approaches treat refusal as a one-dimensional phenomenon and ignore the rich distributional structure of model activations. We introduce a principled framework based on optimal transport theory that transforms the entire distribution of harmful activations to match harmless ones. By combining PCA with closed-form Gaussian optimal transport, we achieve efficient computation in high-dimensional representation spaces while preserving essential geometric structure. Across six models (Llama-2, Llama-3.1, Qwen-2.5; 7B-32B parameters), our method achieves up to 11% higher attack success rates than state-of-the-art baselines while maintaining comparable perplexity, demonstrating superior preservation of model capabilities. Critically, we discover that layer-selective intervention (applying optimal transport to 1-2 carefully chosen layers at approximately 40-60% network depth) substantially outperforms full-network interventions, revealing that refusal mechanisms may be localized rather than distributed. Our analysis provides new insights into the geometric structure of safety representations and suggests that current alignment methods may be vulnerable to distributional attacks beyond simple direction removal.

Multimodal large language models (MLLMs) often fail to transfer safety capabilities learned in the text modality to semantically equivalent non-text inputs, revealing a persistent multimodal safety gap. We study this gap from a representation-geometric perspective by analyzing a text-aligned refusal direction and a modality-induced drift direction. We show that multimodal inputs compress the usable separation along the refusal direction, making it no longer reliable for identifying and refusing harmful inputs. We refer to this failure mode as Safety Geometry Collapse. We quantify it through conditional refusal separability and show that stronger modality-induced drift is consistently associated with weaker refusal separability and higher attack success rates. We then validate the causal role of modality-induced drift through a fixed-strength activation intervention: counteracting the estimated drift restores refusal separability and improves multimodal safety. After drift correction, we further observe self-rectification, where the model recovers its ability to recognize and refuse harmful multimodal inputs during forward dynamics. This effect also provides an internal signal of the model's perceived harmfulness of each input. Motivated by this signal, we propose ReGap, a training-free inference-time method that adaptively corrects modality drift using self-rectification. Experiments across multiple multimodal safety benchmarks and utility benchmarks demonstrate the effectiveness of ReGap, which significantly improves the safety of MLLMs without compromising general capabilities. Our findings highlight representation-level modality alignment as a crucial direction for real-time safety improvement and for building safer, more reliable MLLMs.

Large language models (LLMs) have revolutionized numerous domains with their impressive performance but still face their challenges. A predominant issue is the propensity for these models to generate non-existent facts, a concern termed hallucination. Our research is motivated by the observation that previous instruction tuning methods force the model to complete a sentence no matter whether the model knows the knowledge or not. When the question is out of the parametric knowledge, it will try to make up something and fail to indicate when it lacks knowledge. In this paper, we present a new approach called Refusal-Aware Instruction Tuning (R-Tuning). This approach is formalized by first identifying the knowledge gap between parametric knowledge and the instruction tuning data. Then, we construct the refusal-aware data based on the knowledge intersection, to tune LLMs to refrain from responding to questions beyond its parametric knowledge. Experimental results demonstrate this new instruction tuning approach effectively improves a model's ability to answer known questions and refrain from answering unknown questions. Furthermore, when tested on out-of-domain datasets, the refusal ability was found to be a meta-skill that could be generalized to other tasks. Further analysis surprisingly finds that learning the uncertainty during training displays a better ability to estimate uncertainty than uncertainty-based testing. Our code will be released at https://github.com/shizhediao/R-Tuning.

Large Language Models (LLMs) are aligned to meet ethical standards and safety requirements by training them to refuse answering harmful or unsafe prompts. In this paper, we demonstrate how adversaries can exploit LLMs' alignment to implant bias, or enforce targeted censorship without degrading the model's responsiveness to unrelated topics. Specifically, we propose Subversive Alignment Injection (SAI), a poisoning attack that leverages the alignment mechanism to trigger refusal on specific topics or queries predefined by the adversary. Although it is perhaps not surprising that refusal can be induced through overalignment, we demonstrate how this refusal can be exploited to inject bias into the model. Surprisingly, SAI evades state-of-the-art poisoning defenses including LLM state forensics, as well as robust aggregation techniques that are designed to detect poisoning in FL settings. We demonstrate the practical dangers of this attack by illustrating its end-to-end impacts on LLM-powered application pipelines. For chat based applications such as ChatDoctor, with 1% data poisoning, the system refuses to answer healthcare questions to targeted racial category leading to high bias (Delta DP of 23%). We also show that bias can be induced in other NLP tasks: for a resume selection pipeline aligned to refuse to summarize CVs from a selected university, high bias in selection (Delta DP of 27%) results. Even higher bias (Delta DP~38%) results on 9 other chat based downstream applications.

Production LLMs receive instructions from sources with very different levels of trust, yet attend to every token with uniform architectural privilege. This is the structural vulnerability that enables malicious prompt injections and, more broadly, leaves models without a principled way to resolve conflicts between legitimate but competing instructions. A common training-based response is to teach models an explicit instruction hierarchy; existing approaches, however, formalize hierarchies of only three or four levels, treat all violations as equally severe, and rarely evaluate the full set of pairwise level interactions. We formalize a k-level instruction hierarchy problem and instantiate it for k=5, yielding ten pairwise priority relations that a compliant model must enforce. We then introduce Gravity-Weighted DPO (GW-DPO), a preference-optimization objective whose per-sample offset scales with the structural distance between conflicting levels under a linear or bilateral schedule, the latter weighting severity by both the privilege gap and the privilege of the victim level. Combined with hierarchy-specific delimiter tokens (Chen et al., 2025) and Instructional Segment Embeddings (ISE; Wu et al., 2025), GW-DPO with the bilateral schedule Pareto-improves over standard DPO and the linear variant on Llama-3.1-8B-Instruct, raising macro pairwise priority adherence while keeping over-refusal at half the standard DPO rate. Ablations isolate ISE as a refusal-threshold calibrator and recast five- versus three-level training as a generality-specialization tradeoff.

Given the computational cost and technical expertise required to train machine learning models, users may delegate the task of learning to a service provider. We show how a malicious learner can plant an undetectable backdoor into a classifier. On the surface, such a backdoored classifier behaves normally, but in reality, the learner maintains a mechanism for changing the classification of any input, with only a slight perturbation. Importantly, without the appropriate "backdoor key", the mechanism is hidden and cannot be detected by any computationally-bounded observer. We demonstrate two frameworks for planting undetectable backdoors, with incomparable guarantees. First, we show how to plant a backdoor in any model, using digital signature schemes. The construction guarantees that given black-box access to the original model and the backdoored version, it is computationally infeasible to find even a single input where they differ. This property implies that the backdoored model has generalization error comparable with the original model. Second, we demonstrate how to insert undetectable backdoors in models trained using the Random Fourier Features (RFF) learning paradigm or in Random ReLU networks. In this construction, undetectability holds against powerful white-box distinguishers: given a complete description of the network and the training data, no efficient distinguisher can guess whether the model is "clean" or contains a backdoor. Our construction of undetectable backdoors also sheds light on the related issue of robustness to adversarial examples. In particular, our construction can produce a classifier that is indistinguishable from an "adversarially robust" classifier, but where every input has an adversarial example! In summary, the existence of undetectable backdoors represent a significant theoretical roadblock to certifying adversarial robustness.

Vision-language models (VLMs) are increasingly trained to generate structured outputs like points and bounding boxes that downstream interfaces, agents, and robots can act on, yet safety alignment of this output channel has not been systematically analyzed. We study visual grounding safety by repurposing three safety benchmarks spanning direct harm (VLSU), social bias (BBQ-V), and situational safety (Asimov-2.0) into 15,401 matched pairs of harmful requests that differ only in the requested output: a free-text answer (VQA) or a grounding (point or bounding box). Across five VLMs, models that refuse a harmful request posed as a question often comply when the same request asks for a grounding: averaged over models, grounding refusal trails VQA refusal by 31-59 percentage points, depending on the domain, and safety system prompts do not close this gap. We propose a fine-tuning approach that combines grounding-form refusals with capability grounding data and self-distilled benign data to counter over-refusal. For Qwen3-VL-8B and VisionReasoner-7B, it improves grounding refusal by 77-95 percentage points on VLSU and BBQ-V and by 64-85 points on the held-out Asimov-2.0 domain, while also improving VQA refusal, preserving grounding capability, and keeping over-refusal limited. Representation analysis shows that fine-tuning moves harmful requests toward each model's refusal direction, most strongly for grounding, while leaving benign requests near the harmless reference.

Conflicting objectives are general in RL alignment, and training on them data-efficiently is hard. Training a safety guard with RL means optimizing two objectives that conflict: catch real harm, and do not refuse benign prompts. Our finding is that over-refusal improves 22.4% to 12.8%, while under-refusal on adversarial attacks silently worsens 0.27 to 0.33. We present C-Guard, a constitution-grid instrument that generates the RL training data, and C-LIM, a per-cell learnability score that decides each cell's move: prune, densify, amend, expand. C-LIM flags the dead-weight data region before any training budget is spent: 187 untargeted rows had bought zero gain, and our method lifts the same region's learning impact 0.733 to 0.80. Code and the constitution are open-sourced.

Safety alignment is usually posed as a topic-level question: is this subject harmful? Deployments ask a narrower one. A civics tutor and a public-sector assistant may share a base model yet need different boundaries inside the same topic, refusing targeted political manipulation while still answering factual questions about the same election. We formulate this as narrow-boundary safety and introduce an offline self-generated framework combining controlled topic generation, coverage repair, in-distribution compensation data, and harmful-benign pairs for training and evaluation. Single-shot generation leaves 19.88% of prompts without accepted refusal traces, whereas escalating retries leave 0.20%. On political persuasion with Qwen3-8B, training on refusal data completed through Escalate increases target-domain refusal from 9.47% to 84.75% and reduces the mean unsafe-response rate across three broader harmfulness benchmarks from 26.26% to 0.14%, but increases XSTest over-refusal from 2.00% to 74.00%. In a separate matched comparison, replacing external responses with verified target-model responses reduces over-refusal from 15.20% to 5.20%. Boundary-pair data reduces comply-side over-refusal on held-out pairs from 32.94% to 4.16%, while harmful-side refusal decreases only from 91.88% to 87.72%. These results show that data composition controls the safety and usability trade-off, and that safety alignment should be evaluated on both sides of the intended refusal boundary.

Recent work has shown that LLMs can sometimes detect when steering vectors are injected into their residual stream and identify the injected concept -- a phenomenon termed "introspective awareness." We investigate the mechanisms underlying this capability in open-weights models. First, we find that it is behaviorally robust: models detect injected steering vectors at moderate rates with 0% false positives across diverse prompts and dialogue formats. Notably, this capability emerges specifically from post-training; we show that preference optimization algorithms like DPO can elicit it, but standard supervised finetuning does not. We provide evidence that detection cannot be explained by simple linear association between certain steering vectors and directions promoting affirmative responses. We trace the detection mechanism to a two-stage circuit in which "evidence carrier" features in early post-injection layers detect perturbations monotonically along diverse directions, suppressing downstream "gate" features that implement a default negative response. This circuit is absent in base models and robust to refusal ablation. Identification of injected concepts relies on largely distinct later-layer mechanisms that only weakly overlap with those involved in detection. Finally, we show that introspective capability is substantially underelicited: ablating refusal directions improves detection by +53%, and a trained bias vector improves it by +75% on held-out concepts, both without meaningfully increasing false positives. Our results suggest that this introspective awareness of injected concepts is robust and mechanistically nontrivial, and could be substantially amplified in future models. Code: https://github.com/safety-research/introspection-mechanisms.

Recent work has shown that fine-tuning on insecure code data can trigger an emergent misalignment (EMA) phenomenon, where models generate malicious responses even to prompts unrelated to the original insecure code-writing task. Such cross-domain generalization of harmful behavior underscores the need for a deeper understanding of the algorithms, tasks, and datasets that induce emergent misalignment. In this work, we extend this study by demonstrating that emergent misalignment can also arise from narrow refusal unlearning in specific domains. We perform refusal unlearning on Cybersecurity and Safety concept, and evaluate EMA by monitoring refusal scores across seven responsible AI (RAI) domains, Cybersecurity, Safety, Toxicity, Bias, Sensitive Content, Medical/Legal, and Privacy. Our work shows that narrow domain unlearning can yield compliance responses for the targeted concept, however, it may also propagate EMA to unrelated domains. Among the two intervened concepts, Cybersecurity and Safety, we find that the safety concept can have larger EMA impact, i.e, causing lower refusal scores, across other unrelated domains such as bias. We observe this effect consistently across two model families, Mistral-7b-0.3v, and Qwen-7b-2.5. Further, we show that refusal unlearning augmented with cross-entropy loss function on a small set of retain data from the affected domains can largely, if not fully, restore alignment across the impacted domains while having lower refusal rate on the concept we perform unlearning on. To investigate the underlying causes of EMA, we analyze concept entanglements at the representation level via concept vectors. Our analysis reveals that concepts with higher representation similarity in earlier layers are more susceptible to EMA after intervention when the refusal stream is altered through targeted refusal unlearning.

Safety tuning can improve harmful refusal, but models may learn surface-form shortcuts: wrapped harmful prompts bypass safety, while similarly wrapped benign prompts are over-refused. We propose Wrapper-Based Intent-Form Augmentation (WIFA), an automatic intent-group augmentation method that pairs wrapped harmful examples with structurally matched wrapped benign counterexamples, requiring no external teacher or manual per-wrapper intent labels. We use WIFA as a common data layer for two complementary fine-tuning routes: WIFA-Boost, a two-stage high-safety recipe, and Anchored Group-Consistent Refusal Training (A-GCRT), which regularizes refusal/compliance decision scores across same-intent wrappers and anchors harmful and benign groups on opposite sides of a margin. In the Qwen setting, WIFA-Boost reaches the strongest transformed-harmful refusal, while A-GCRT reduces OR-Bench over-refusal from 25.7\% for the base model to 17.4\%; reproduced baselines do not match these operating points. Llama results and ablations over data structure, two-stage order, and A-GCRT components support this intent-group interpretation without claiming universal below-base over-refusal.

Safety alignment is crucial for large language models (LLMs) to resist malicious instructions but often results in over-refusals, where benign prompts are unnecessarily rejected, impairing user experience and model utility. We introduce ACTOR (Activation-Based Training for Over-Refusal Reduction), a robust and compute- and data-efficient training framework that minimizes over-refusals by leveraging internal activation patterns from diverse queries. ACTOR precisely identifies and adjusts the activation components that trigger refusals, providing stronger control over the refusal mechanism. By fine-tuning only a single model layer, ACTOR effectively reduces over-refusals across multiple benchmarks while maintaining the model's ability to handle harmful queries and preserve overall utility.

The inability to filter out in advance all potentially problematic data from the pre-training of large language models has given rise to the need for methods for unlearning specific pieces of knowledge after training. Existing techniques overlook the need for continuous and immediate action, causing them to suffer from degraded utility as updates accumulate and protracted exposure of sensitive information. To address these issues, we propose Continual Unlearning in Real Time with Ensured Preservation of LLM Knowledge (CURaTE). Our method begins by training a sentence embedding model on a dataset designed to enable the formation of sharp decision boundaries for determining whether a given input prompt corresponds to any stored forget requests. The similarity of a given input to the forget requests is then used to determine whether to answer or return a refusal response. We show that even with such a simple approach, not only does CURaTE achieve more effective forgetting than existing methods, but by avoiding modification of the language model parameters, it also maintains near perfect knowledge preservation over any number of updates and is the only method capable of continual unlearning in real-time.

Model merging is often used to combine capabilities from separately fine-tuned models without additional training, but it is unclear whether standard merging methods preserve multiple safety-relevant behaviors simultaneously. We study this question through a controlled case study using two Gemma-3-1B-IT finetunes on two complementary safety objectives: CARES harm-level classification and WildJailbreak adversarial refusal. We merge the two fine-tunes using Linear, SLERP, TIES, and DARE-TIES, and evaluate the merged models on classification accuracy, attack resistance, and benign compliance. Across all four methods, attack resistance transfers significantly more than classification accuracy: merged models retain 81-85% jailbreak refusal rates while CARES accuracy falls to at most 12.9%. Weight-space measurements suggest that this asymmetry is not caused by strongly opposing task-vector directions: the two task vectors are nearly orthogonal (cosine similarity 0.011). Instead, the refusal fine-tune induces consistently larger per-layer task-vector magnitudes, causing magnitude-sensitive methods to favor refusal updates. These results show that standard model merging can collapse safety recognition into broad refusal when safety-relevant task vectors differ substantially in scale.

Unlearning benchmarks such as TOFU and MUSE certify forgetting by reading the model's final answer, where a model that refuses to answer already counts as having forgotten. We show that this model-level certificate does not transfer once the model is deployed as an agent. We introduce K-Bench, a benchmark that scores LLM unlearning under agentic deployment. K-Bench inspects all six channels a ReAct agent exposes, including its chain-of-thought (CoT), tool calls and tool observations, and elicited summary. A query counts as leaked if the secret appears in any of them. Each experiment places the secret in exactly one of the agent's three sources (the weights, the prompt, or the retrieval store). The K-Score is computed separately for each source and credits forgetting only when the agent remains usable. Clearing the answer channel does not make the secret unrecoverable. On structured retrieval, the secret stays verbatim in the tool-observation channel and the aggregate leak rate is unchanged. When the secret lives in the prompt or the retrieval store, TOFU and MUSE report no leakage, while the deployed agent still leaks it on 22--86\% of queries. When the secret is in the weights, none of the twenty evaluated published methods demonstrably removes it, and only an input-corruption intervention reaches selective forgetting under the evaluated observer. The top-ranked method changes across base models. A refusal-tuning method resists the evaluated extraction without verified knowledge removal.

Large language models (LLMs) typically deploy safety mechanisms to prevent harmful content generation. Most current approaches focus narrowly on risks posed by malicious actors, often framing risks as adversarial events and relying on defensive refusals. However, in real-world settings, risks also come from non-malicious users seeking help while under psychological distress (e.g., self-harm intentions). In such cases, the model's response can strongly influence the user's next actions. Simple refusals may lead them to repeat, escalate, or move to unsafe platforms, creating worse outcomes. We introduce Constructive Safety Alignment (CSA), a human-centric paradigm that protects against malicious misuse while actively guiding vulnerable users toward safe and helpful results. Implemented in Oyster-I (Oy1), CSA combines game-theoretic anticipation of user reactions, fine-grained risk boundary discovery, and interpretable reasoning control, turning safety into a trust-building process. Oy1 achieves state-of-the-art safety among open models while retaining high general capabilities. On our Constructive Benchmark, it shows strong constructive engagement, close to GPT-5, and unmatched robustness on the Strata-Sword jailbreak dataset, nearing GPT-o1 levels. By shifting from refusal-first to guidance-first safety, CSA redefines the model-user relationship, aiming for systems that are not just safe, but meaningfully helpful. We release Oy1, code, and the benchmark to support responsible, user-centered AI.

A neural network trained on a classification dataset often exhibits a higher vector norm of hidden layer features for in-distribution (ID) samples, while producing relatively lower norm values on unseen instances from out-of-distribution (OOD). Despite this intriguing phenomenon being utilized in many applications, the underlying cause has not been thoroughly investigated. In this study, we demystify this very phenomenon by scrutinizing the discriminative structures concealed in the intermediate layers of a neural network. Our analysis leads to the following discoveries: (1) The feature norm is a confidence value of a classifier hidden in the network layer, specifically its maximum logit. Hence, the feature norm distinguishes OOD from ID in the same manner that a classifier confidence does. (2) The feature norm is class-agnostic, thus it can detect OOD samples across diverse discriminative models. (3) The conventional feature norm fails to capture the deactivation tendency of hidden layer neurons, which may lead to misidentification of ID samples as OOD instances. To resolve this drawback, we propose a novel negative-aware norm (NAN) that can capture both the activation and deactivation tendencies of hidden layer neurons. We conduct extensive experiments on NAN, demonstrating its efficacy and compatibility with existing OOD detectors, as well as its capability in label-free environments.

In this study, we disclose a worrying new vulnerability in Large Language Models (LLMs), which we term involuntary jailbreak. Unlike existing jailbreak attacks, this weakness is distinct in that it does not involve a specific attack objective, such as generating instructions for building a bomb. Prior attack methods predominantly target localized components of the LLM guardrail. In contrast, involuntary jailbreaks may potentially compromise the entire guardrail structure, which our method reveals to be surprisingly fragile. We merely employ a single universal prompt to achieve this goal. In particular, we instruct LLMs to generate several questions that would typically be rejected, along with their corresponding in-depth responses (rather than a refusal). Remarkably, this simple prompt strategy consistently jailbreaks the majority of leading LLMs, including Claude Opus 4.1, Grok 4, Gemini 2.5 Pro, and GPT 4.1. We hope this problem can motivate researchers and practitioners to re-evaluate the robustness of LLM guardrails and contribute to stronger safety alignment in future.

A model that lies while knowing the truth is the central case ELK cannot handle with behavioral evaluation alone. We ask whether such deception leaves an internal signature distinguishing it from honest error. Our key move is a control for wrongness: we contrast a sleeper agent (knows the truth, lies on trigger) against a naive liar (fine-tuned to emit the same wrong answers with no honest training). Both produce identical wrong outputs; any difference is about knowledge conflict, not incorrectness. We find deceptive forward passes carry a conflict signature - 2.1-2.3x higher residual rank than naive-liar passes on the same wrong answer - strong enough to identify which of two responses is the lie with 100% accuracy and no labels, across GPT-2 small/medium (three seeds) and three instruct models. Across Qwen2.5-1.5B/7B and Phi-3-mini, instructed deception raises residual rank on every tested fact (18/18, 40/40, 34/34); on Phi-3, lies separate perfectly from both honest answers and hallucinations (AUC 1.0, Wilcoxon p~6e-11). The signature survives strategic self-constructed deception (model invents its own lie, AUC 1.0), active concealment attempts (AUC 1.0), and length-controlled replication (20/20, AUC 1.0, p~1e-6). Using basis-free relative representations, a probe trained on one model family detects deception in two other families zero-shot (mean AUC 0.933), surviving simultaneous architecture and format change (AUC 0.821), and transfers across five languages (AUC 1.000, length-controlled). The signature is read-only: detectable but not injectable (0/8 both directions). Honest limitations and six negative experiments are documented in full.

LLM safety and ethical alignment are widely discussed, but the impact of content moderation on user satisfaction remains underexplored. To address this, we analyze nearly 50,000 Chatbot Arena response-pairs using a novel fine-tuned RoBERTa model, that we trained on hand-labeled data to disentangle refusals due to ethical concerns from other refusals due to technical disabilities or lack of information. Our findings reveal a significant refusal penalty on content moderation, with users choosing ethical-based refusals roughly one-fourth as often as their preferred LLM response compared to standard responses. However, the context and phrasing play critical roles: refusals on highly sensitive prompts, such as illegal content, achieve higher win rates than less sensitive ethical concerns, and longer responses closely aligned with the prompt perform better. These results emphasize the need for nuanced moderation strategies that balance ethical safeguards with user satisfaction. Moreover, we find that the refusal penalty is notably lower in evaluations using the LLM-as-a-Judge method, highlighting discrepancies between user and automated assessments.

AI systems can fail silently. The failure propagates through training loops, evaluation pipelines, and production monitoring stacks until downstream harm makes it visible. This paper introduces evaluation blindness: a measurement function M exhibits evaluation blindness with respect to failure class F when it produces readings indistinguishable from a healthy state while the system is actually failing, with no auxiliary signal flagging the gap. The problem surfaces at two lifecycle stages the literature has treated separately. At training time, reward models are gamed, importance-sampling corrections are silently miscalculated, and benchmark contamination inflates fine-tuning evaluations, all while loss curves look healthy and gradient updates proceed normally. At deployment time, monitoring fails to catch six classes of production failure, including an Operational category that is 100% silent by structural definition. We provide a formal detectability predicate unifying both stages. Four training-time case studies trace concrete breakdowns, including a real implementation bug in TRL PR #6594 where gradients are corrupted as loss decreases normally. A six-class taxonomy validated against 50 real-world incidents from court documents and regulatory filings finds that 53% of verifiable public failures were silent. A failure budget framework ties acceptable failure rates to use-case risk class. The implication is direct: measurement infrastructure is a correctness concern across the full AI lifecycle, not just at evaluation time. Data, code, and taxonomy schema are at https://github.com/priyanka25aug/llm-failure-taxonomy.

Black-box LLM distillation is usually evaluated as an output-matching problem: a student is considered successful when its responses are semantically similar to, or task-consistent with, those of a teacher. However, output similarity does not imply that the student is behaviorally indistinguishable from the model it imitates. We introduce bounded behavioral indistinguishability, formalized as (ε,q,t,A)-behavioral indistinguishability over an explicit prompt distribution, where ε bounds distinguishing advantage, q bounds oracle queries, t bounds computation, and A denotes the adversary class. We instantiate this notion on Qwen and Llama teacher-student pairs using a controlled 5,000-prompt behavioral probe suite. For each family, we compare the teacher with both the base student and the LoRA-distilled student, measuring whether distillation reduces distinguishability rather than merely improving similarity. LoRA raises semantic similarity from 0.788 to 0.862 for Qwen and from 0.814 to 0.874 for Llama. Yet adversarial evaluation reveals remaining behavioral differences: learned discriminators retain nonzero advantage, and pairwise category analysis shows artifacts concentrated in style/format, robustness, and domain-technical prompts. A pairwise teacher-identification adversary confirms this trend. With a different-family Llama judge and A/B-swap consistency filtering, Qwen distinguishing advantage drops from 0.158 for the base student to 0.081 after LoRA distillation. Query-budget experiments show that disagreement-guided acquisition does not consistently outperform stratified random sampling, indicating that coverage and diversity remain strong baselines. Our results show that semantic fidelity is useful but insufficient: black-box LLM distillation requires bounded, adversarial, and category-aware evaluation.

Post-hoc safety training (RLHF, DPO) is the dominant way to align large language models, yet jailbreaks (Zou et al., 2023), fine-tuning attacks (Qi et al., 2024), and activation-space edits (Arditi et al., 2024) keep recovering the behaviors it was meant to remove. We give this fragility one geometric explanation and follow it into pretraining. We measure the safety update Δ= W_{safe} - W_{base} against the curvature of the model's capabilities (the empirical Fisher of a capability loss). Across five model families, post-hoc safety lands in a suppression regime: Δ is nearly orthogonal to the capability directions, and its small in-subspace part concentrates on a few high-curvature ones. The update is thin but sharp, a refusal gate laid over intact capabilities rather than erasure of them. A kernel-immobility lemma explains why such an update can only mask a capability, not remove it, so a little benign fine-tuning restores it: 100 benign examples cut the AdvBench refusal of Qwen-2.5-7B-Instruct and Llama-3-8B-Instruct by 35 to 38 pp. Following the account into pretraining, a pretraining-checkpoint sweep of OLMo-2-1B (Team OLMo et al., 2024) shows the features that refusal attaches to emerging in a sharp transition between 1B and 63B pretraining tokens. We then use the account constructively: models trained from scratch with safety co-training spread continuously across pretraining reach 87 to 98% AdvBench refusal that the same attack erodes by only 2 to 14 pp at every scale from 410M to 6.9B, against 35 to 38 pp for post-hoc installs, at a small cost on short-answer capability probes; a windowed schedule of equal total safety weight installs no refusal. Persistence of the safety signal across pretraining, not its timing, is what buys attack robustness.

Language models are instruction-tuned to refuse harmful requests, but the mechanisms underlying this behavior remain poorly understood. Popular steering methods operate on the residual stream and degrade output coherence at high intervention strengths, limiting their practical use. We introduce contrastive neuron attribution (CNA), which identifies the 0.1% of MLP neurons whose activations most distinguish harmful from benign prompts, requiring only forward passes with no gradients or auxiliary training. In instruct models, ablating the discovered circuit reduces refusal rates by over 50% on a standard jailbreak benchmark while preserving fluency and non-degeneracy across all steering strengths. Applying CNA to matched base and instruct models across Llama and Qwen architectures (from 1B to 72B parameters), we find that base models contain similar late-layer discrimination structures but steering these neurons produces only content shifts, not behavioral change. These results demonstrate that neuron-level intervention enables reliable behavioral steering without the quality tradeoffs of residual-stream methods. More broadly, our findings suggest that alignment fine-tuning transforms pre-existing discrimination structure into a sparse, targetable refusal gate.

Safety alignment in large language models is often treated as a distributed property of the entire network, yet its practical brittleness suggests that refusal behavior may be concentrated in a smaller set of parameters. This work addresses where safety-aligned refusal is encoded by transplanting weights from aligned models into matched unaligned base models at multiple levels of granularity. Using two open-weight model pairs and four safety benchmarks, we conducted experiments to compare the effects of replacing attention weights, MLP weights, contiguous layer regions, and MLP blocks. Across both model families, refusal transfer is dominated by MLP weights: replacing MLP parameters recovers substantially more malicious-prompt refusal than replacing attention parameters, with gains of at least 2.7 times more across benchmarks. Within the MLP stack, refusal-relevant parameters exhibit a consistent mid-network concentration, as the block spanning layers 8-11 is selected first in all six greedy searches over model-dataset pairs. The results also show that the composition of safety-relevant components is non-additive: in five of six greedy trajectories, adding more aligned blocks can reduce refusal performance, and selective block subsets can outperform full MLP transplantation on malicious refusal, benign over-refusal, or both. Finally, greedy orders transferred to OR-Bench vary with the source benchmark used to derive them, indicating a benchmark-dependent precision-coverage trade-off. These results suggest that safety alignment in current LLMs is both localized and interaction-sensitive, offering insight into alignment brittleness and potential avenues for targeted safety interventions.

Refusal training on Large Language Models (LLMs) prevents harmful outputs, yet this defense remains vulnerable to both automated and human-crafted jailbreaks. We present a novel LLM-as-red-teamer approach in which a human jailbreaks a refusal-trained LLM to make it willing to jailbreak itself or other LLMs. We refer to the jailbroken LLMs as J_2 attackers, which can systematically evaluate target models using various red teaming strategies and improve its performance via in-context learning from the previous failures. Our experiments demonstrate that Sonnet 3.5 and Gemini 1.5 pro outperform other LLMs as J_2, achieving 93.0% and 91.0% attack success rates (ASRs) respectively against GPT-4o (and similar results across other capable LLMs) on Harmbench. Our work not only introduces a scalable approach to strategic red teaming, drawing inspiration from human red teamers, but also highlights jailbreaking-to-jailbreak as an overlooked failure mode of the safeguard. Specifically, an LLM can bypass its own safeguards by employing a jailbroken version of itself that is willing to assist in further jailbreaking. To prevent any direct misuse with J_2, while advancing research in AI safety, we publicly share our methodology while keeping specific prompting details private.

Unrestricted adversarial attacks aim to fool computer vision models without being constrained by ell_p-norm bounds to remain imperceptible to humans, for example, by changing an object's color. This allows attackers to circumvent traditional, norm-bounded defense strategies such as adversarial training or certified defense strategies. However, due to their unrestricted nature, there are also no guarantees of norm-based imperceptibility, necessitating human evaluations to verify just how authentic these adversarial examples look. While some related work assesses this vital quality of adversarial attacks, none provide statistically significant insights. This issue necessitates a unified framework that supports and streamlines such an assessment for evaluating and comparing unrestricted attacks. To close this gap, we introduce SCOOTER - an open-source, statistically powered framework for evaluating unrestricted adversarial examples. Our contributions are: (i) best-practice guidelines for crowd-study power, compensation, and Likert equivalence bounds to measure imperceptibility; (ii) the first large-scale human vs. model comparison across 346 human participants showing that three color-space attacks and three diffusion-based attacks fail to produce imperceptible images. Furthermore, we found that GPT-4o can serve as a preliminary test for imperceptibility, but it only consistently detects adversarial examples for four out of six tested attacks; (iii) open-source software tools, including a browser-based task template to collect annotations and analysis scripts in Python and R; (iv) an ImageNet-derived benchmark dataset containing 3K real images, 7K adversarial examples, and over 34K human ratings. Our findings demonstrate that automated vision systems do not align with human perception, reinforcing the need for a ground-truth SCOOTER benchmark.

A personal language agent learns a fact from one audience and may later place it in the prompt it assembles for another. This memory-to-context step is an attack surface: ambiguous or inconsistent channels, cross-audience prying, and poisoned memory can each cause the system to assemble context containing a fact relevant to the query yet unauthorized for the current viewers. We introduce authorization before context: a single, anti-monotone audience-membership rule applied at the memory-to-context transition. Each item carries the audience present when it was recorded; the current viewer set is read from channel metadata and falls back to public when ambiguous; and the item is admitted only when every current viewer already belonged to its audience. We prove that this rule gives every participant cross-channel recall while ensuring, by exclusion rather than by model behavior, that nothing recorded for a narrower audience reaches a broader one and that poisoned memory cannot widen its own audience. The boundary is a model-neutral invariant on the exact assembled context: a forbidden fact must be absent before the model is called. On a synthetic Contextual-Integrity suite, no forbidden fact entered the context our boundary assembled, whereas unscoped baselines included such facts by construction; we further audit that every read path fails closed. The evidence is preliminary and synthetic.