In this work, we show that natural policy gradient, a core algorithm in reinforcement learning, admits an exact formulation as a smoothed and averaged form of policy iteration. Specifically, we introduce doubly smoothed policy iteration (DSPI), a Bellman-operator framework in which each policy is obtained by applying a regularized greedy step to a weighted average of past $Q$-functions. DSPI includes policy iteration, dual-averaged policy iteration, natural policy gradient, and more general policy dual averaging methods as special cases. Using only monotonicity and contraction of smoothed Bellman operators, we prove distribution-free global geometric convergence of DSPI. Consequently, standard natural policy gradient and policy dual averaging achieve an iteration complexity of $\mathcal{O}((1-γ)^{-1}\log((1-γ)^{-1}ε^{-1}))$ for computing an $ε$-optimal policy, without modifying the MDP, adding regularization beyond the mirror map inherent in the update, or using adaptive, trajectory-dependent stepsizes. For the unregularized greedy case, corresponding to dual-averaged policy iteration, we also prove finite termination. The same Bellman-operator framework further extends to discounted MDPs with linear function approximation and stochastic shortest path problems.

We propose a closed-form spectral framework for relative log-density estimation in linearly parameterized probabilistic models, including unnormalized and conditional models. This is achieved by representing the Kullback-Leibler (KL) divergence as an integral of weighted chi-squared divergences, converting KL estimation into a family of least-squares problems. We derive an explicit spectral formula based only on first- and second-order feature moments, yielding closed-form estimators of both divergences and log-density potentials for fixed features. The framework extends to a broad class of f-divergences and can be combined with kernelization or feature learning with neural networks. We prove convergence guarantees for the resulting estimators and empirically compare them on synthetic data with optimization-based variational formulations, including logistic and softmax regression for normalized conditional models.

Experience-driven self-evolving agents aim to overcome the static nature of large language models by distilling reusable experience from past interactions, thus enabling adaptation to novel tasks at deployment time. This process places substantial demands on the foundation model's capacities for abstraction, generalization, and in-context learning. However, most existing studies focus primarily on system-level design choices, such as how experience is represented and managed, neglecting the inherent capabilities of the underlying model. While some recent works have started to optimize the experience utilization stage via reinforcement learning, they still fail to treat self-evolution as a unified process to be jointly optimized. To this end, we propose Evolving-RL, an efficient algorithmic framework that jointly improves the experience extraction and utilization capabilities required for self-evolution. Specifically, we center the learning process on experience extraction and evaluation, using the two supervisory signals derived from evaluation to optimize the extractor and solver separately and thus enable their coordinated co-evolution. Experiments on ALFWorld and Mind2Web show that Evolving-RL effectively enhances LLMs' ability to extract and reuse experience, leading to strong performance gains on out-of-distribution tasks (up to 98.7% relative improvement over the GRPO baseline on ALFWorld unseen tasks and 35.8% on Mind2Web), and these gains are fully unlocked only through the coordinated co-evolution of experience extraction and utilization. Furthermore, Evolving-RL inherently functions as an experience-augmented RL algorithm. By internalizing reusable experience patterns directly into model parameters, it achieves remarkable performance gains over standard baselines on both seen and unseen tasks, even in the absence of test-time experience accumulation.

Vision Transformers (ViTs) are built by stacking independently parameterized blocks, but it remains unclear how much of this depth requires layer specific transformations and how much can be realized through recurrent computation. We study this question with bViT, a single-block recurrent ViT in which one transformer block is applied repeatedly to process an image. This architecture preserves the iterative structure of a deep ViT while removing layer specific block parameterization, providing a controlled setting for studying recurrence in vision. On ImageNet-1K, a 12-step bViT-B achieves accuracy comparable to standard ViT-B under the same training recipe and computational budget, while using an order of magnitude fewer parameters. We observe that recurrent performance improves with representation width, with wider bViTs recovering much more of the performance of standard ViTs than narrow variants. We interpret this behavior as implicit depth multiplexing, where a shared block expresses multiple step-dependent computations through the evolving hidden state. Beyond ImageNet classification, bViT transfers competitively to downstream tasks and enables parameter-efficient fine-tuning. Mechanistic analyses of activations, attention and step-specific pruning show that the shared block changes its effective behavior across recurrent steps rather than simply repeating the same computation. Our results suggest that a large fraction of ViT depth can be implemented through recurrent reuse, provided that the representation space is sufficiently wide.

Digital personas powered by Large Language Models (LLMs) are increasingly proposed as substitutes for human survey respondents, yet it remains unclear when they can reliably approximate human survey findings. We answer this question using the LISS panel, constructing personas from respondents' background variables and pre-2023 survey histories, then testing them against the same respondents' held-out post-cutoff answers. Across four persona architectures, three LLMs, and two prediction tasks, we assess performance at the question, respondent, distributional, equity, and clustering levels. Digital personas improve alignment with human response distributions, especially in domains tied to stable attributes and values, but remain limited for individual prediction and fail to recover multivariate respondent structure. Retrieval-augmented architectures provide the clearest gains, but performance depends more on human response structure than on model choice: personas perform best for low-variability questions and common respondent patterns, and worst for subjective, heterogeneous, or rare responses. Our results provide practical guidance on when digital personas could be appropriate for survey research and when human validation remains necessary.

Trellis-coded quantization sets the current 2-bit post-training frontier for LLMs (QTIP), but pushing below the PTQ ceiling requires quantization-aware training, and QAT on a trellis is obstructed by the non-differentiable Viterbi argmax. We introduce BCJR-QAT, a relaxation that replaces the argmax with the BCJR forward-backward sum-product algorithm at temperature $T$, producing a soft codeword equal to the Boltzmann expectation over trellis paths, exactly differentiable, recovering the hard QTIP code as $T \to 0$, and mathematically identical to the transfer-matrix computation for a 1D Ising-like spin chain. We contribute (i) a fused Triton kernel making BCJR tractable on a single consumer GPU ($6.57\times$ speedup, fp32 parity); (ii) a quantitative drift-budget theory of when BCJR-QAT can escape the QTIP-PTQ Voronoi basin, verified across four experiments; and (iii) a positive empirical result on Llama-3.2-1B at 2 bpw under end-to-end forward-KL distillation: with the right schedule (skip the high-$T$ phase to avoid an overshoot we diagnose), single-layer BCJR-QAT beats QTIP-PTQ by $\mathbf{-0.084}$ PPL on WikiText-2, and multi-layer compounding is super-additive.

We consider the active learning problem where the goal is to learn an unknown function with low prediction error under an unknown Boltzmann distribution induced by the function itself. This self-induced weighting arises naturally in problems such as potential energy surface (PES) modeling in computational chemistry, yet poses unique challenges as the target distribution is unknown and its partition function is intractable. We propose \texttt{AB-SID-iVAR}, a Gaussian Process-based acquisition function that approximates the intractable Bayesian target distribution in closed form while avoiding partition function estimation, and is applicable to both discrete and continuous input domains. We also analyze a Thompson sampling alternative (\texttt{TS-SID-iVAR}) as a higher variance Monte Carlo variant. Despite the unknown target, under mild conditions, we establish that the terminal prediction error vanishes with high probability, and provide a tighter average-case guarantee. We demonstrate consistent improvements over existing approaches in this setting on synthetic benchmarks and real-world PES modeling and drug discovery tasks.

Constraint-based causal discovery is widely used for learning causal structures, but heavy reliance on conditional independence (CI) testing makes it computationally expensive in high-dimensional settings. To mitigate this limitation, many divide-and-conquer frameworks have been proposed, but most assume causal sufficiency, i.e., no latent variables. In this paper, we show that divide-and-conquer strategies can be theoretically generalized beyond causal sufficiency to settings with latent variables. Specifically, we propose a recursive decomposition framework, termed DiCoLa, that enables divide-and-conquer causal discovery in the presence of latent variables. It recursively decomposes the global learning task into smaller subproblems and integrates their solutions through a principled reconstruction step to recover the global structure. We theoretically establish the soundness and completeness of the proposed framework. Extensive experiments on synthetic data demonstrate that our approach significantly improves computational efficiency across a range of causal discovery algorithms, while experiments on a real-world dataset further illustrate its practical effectiveness.

Proper weight initialization prior to training has historically been one of the key factors that helped kick off the deep learning revolution. Initialization is even more crucial in "reservoir computing", where the weights of a readout layer are learned linearly while the reservoir weights are fixed and largely determine the richness, stability and memory of the resulting dynamics. In the infinite-width limit it has been shown that meaningful initializations are those sitting at an effective critical point of the randomly initialized model. The phase transition is controlled by the weight variance $g^2$ and separates an ordered phase from a chaotic one where information progressively degrades. Here we derive a simple criterion to estimate the critical $g_c$ for a broad class of recurrent architectures and we show that it closely tracks the gain at which a gated-RNN reservoir achieves peak performance on a chaotic forecasting task. Finally, we argue that our criterion can serve as a design principle for future initialization schemes.

Trajectories are nowadays valuable information for a wide range of applications. However they are also inherently sensitive, as they contain highly personal information about individuals. Facing this challenge, synthesizing mobility trajectories has emerged as a promising solution to leverage mobility information while preserving privacy. State-of-the-art models, often rely on the false assumptions of generative models implicit privacy and fails to provide privacy guarantees while preserving trajectories utility. Here, we introduce diffGHOST, a conditional diffusion model based on latent space segmentation, designed to answer this challenge. Thus, this paper propose a methodology that identify and mitigate memorization of critical samples using condition segments of a learn latent space.

Data-driven medical AI is traditionally formulated as a discriminative mapping from input $X$ to output $Y$ via a learned function $f$, which does not generalize well across heterogeneous data and modalities encountered in real-world clinical settings. In this work, we propose a fundamentally different, generative paradigm. We model the joint distribution $P(X,Y)$ using diffusion models and reframe inference as a test-time output optimization problem. By guiding the generative process to match observed inputs, our framework enables flexible, gradient-based conditioning at inference time without architectural changes or retraining, effectively supporting arbitrary and previously unseen combinations of observations. Extensive experiments demonstrate strong performance across standard and cross-modality medical image segmentation, few-shot segmentation with only 2 or 4 training samples, degraded-input segmentation, shape completion from sparse and partial observations, and zero-shot application to demonstrate generality. To support these evaluations, we curated and released a large-scale text-shape dataset derived from MedShapeNet. Our results highlight the versatility of generative joint modeling as a foundation for reusable, task-agnostic medical AI systems.

Modern language models scale depth by stacking layers, each holding its own state - a per-layer KV cache in transformers, a per-layer matrix in Mamba, Gated DeltaNet (GDN), RWKV, and xLSTM. Biological systems lean heavily on recurrence rather than on stacking. We ask how far that shape can go on language modeling. We propose Grounded Prediction Networks (GPN): one state vector revisited at every step through a single recurrent block - one FFN, one shared matrix memory. At 130M parameters, a 1-layer GPN+M reaches FineWeb-Edu perplexity 18.06, within 13% of a 12-layer Transformer++ (16.05) and 18% of a 10-layer GDN (15.34); a 2-layer variant closes the gap to 6%/11%. We do not match the deep baselines. Because the working context is a single vector, we can directly inspect its geometry: a persistent default-token direction, a content-bearing horizon of tens of tokens, and memory heads that split spontaneously into fast and slow retention pools.

Pretrained diffusion models provide powerful learned priors, but in scientific sampling the target distribution often depends on physical context that is not fully represented by one generative model. We introduce Generative Gibbs for Physics-Aware Sampling (GG-PA), a training-free framework that formulates the composition of learned partial priors and explicit physical context as inference over a joint target distribution in an augmented state space. We derive a Gibbs sampler for this joint target, show that it is asymptotically exact as the diffusion time approaches zero, and prove that in settings with quadratic interactions it remains exact at finite diffusion times. We further introduce replica exchange over diffusion time to accelerate mixing. Experiments on a double-well system, a $ϕ^4$ lattice model, and atomistic peptide systems show that GG-PA recovers context-induced distribution shifts and emergent collective behavior in interacting systems using partial priors without retraining. These results demonstrate GG-PA as a practical approach for combining pretrained generative priors with explicit physical context.

Large Vision-Language Models (VLMs) are successful in addressing a multitude of vision-language understanding tasks, such as Visual Question Answering (VQA), but their memory and compute requirements remain a concern for practical deployment. A promising class of techniques for mitigating this concern is Knowledge Distillation, where knowledge from a high-capacity Teacher network is transferred to a considerably smaller Student network. However, the capacity gap between the two networks is both a blessing and a curse: the smaller the Student network, the better its efficiency, and the larger the Teacher, the more knowledge it carries; yet, beyond a point, the larger capacity gap between the two leads to worse knowledge transfer. To counter this effect, we propose a bottom-up cascaded knowledge distillation (CKD) framework. Instead of treating knowledge transfer as an activity involving one high-capacity Teacher (or an ensemble of such), inspired by human formal education systems, we introduce one (potentially, more) additional Teacher(s) of intermediate capacity that gradually bring the Student network to the next level, where the next (higher-capacity) Teacher can take over. We provide a theoretical analysis in order to study the effect of cascaded distillation in the generalization performance of the Student. We apply the proposed framework on models build upon the LLaVA methodology and evaluate the derived models on seven standard, publicly available VQA benchmarks, demonstrating their SotA performance.

Continual Pre-Training (CPT) is essential for enabling Language Models (LMs) to integrate new knowledge without erasing old. While classical CPT techniques like data replay have become the standard paradigm, the mechanisms underlying how LMs acquire and retain facts over time, termed as continual Factual Knowledge Acquisition (cFKA), remain unclear. In this work, we present a theoretical framework that characterizes the training dynamics of cFKA using a single-layer Transformer, offering a unified explanation for the behavior of representative CPT methods. Our analysis reveals that regularization-based methods merely adjust the convergence rate of parameters without altering the inherent forgetting tendency, whereas data replay methods succeed in shifting convergence dynamics and stabilizing pretrained knowledge. Building on these insights, we propose a novel generative data replay approach, called \textbf{S}electing \textbf{T}okens via attenti\textbf{O}n \textbf{C}ontribution~(STOC), which identifies influential factual snippets to guide replay data generation. Extensive experiments on both synthetic and real-world datasets validate our findings and demonstrate that STOC effectively enhances cFKA by mitigating catastrophic forgetting.

The rapid adoption of LLMs in both research and industry highlights the challenges of deploying them safely and reveals a gap in the systematic evaluation of toxicity benchmarks. As organizations increasingly rely on these benchmarks to certify models for customer-facing applications and automated moderation, unrecognized evaluation biases could lead to the deployment of vulnerable or unsafe systems. This work investigates the robustness of established benchmarking setups and examines how to measure currently neglected intrinsic biases, such as those related to model choice, metrics, and task types. Our experiments uncover significant discrepancies in benchmark behaviors when evaluation setups are altered. Specifically, shifting the task from text completion to summarization increases the tendency of benchmarks to flag content as harmful. Additionally, certain benchmarks fail to maintain consistent behavior when the input data domain is changed. Furthermore, we observe model-specific instabilities, demonstrating a clear need for more robust and comprehensive safety evaluation frameworks.

LLM-based automatic heuristic design has shown promise for generating executable heuristics for combinatorial optimization, but existing methods mainly rely on delayed endpoint performance. We propose a \emph{teacher-aware evolutionary framework} that uses independently trained learned optimization policies as behavioral teachers. Instead of deploying or imitating the teacher, our method queries it on states visited by candidate heuristic programs and uses its action preferences as local feedback for evolution. The resulting search discovers static executable heuristics guided by both task performance and teacher-derived behavioral signals. Experiments on scheduling, routing, and graph optimization benchmarks show that our method improves over performance-driven LLM heuristic evolution baselines while requiring no neural inference at deployment. These results suggest that learned optimization policies can be repurposed as behavioral feedback sources for automatic heuristic discovery.

Fine-tuning Large Language Models (LLMs) on benign narrow data can sometimes induce broad harmful behaviors, a vulnerability termed emergent misalignment (EM). While prior work links these failures to specific directions in the activation space, their relationship to the model's broader persona remains unexplored. We map the latent personality space of LLMs through established psychometric profiles like the Big Five, Dark Triad, and LLM-specific behaviors (e.g. evil, sycophancy), and show that the semantic geometry is highly stable across aligned models and their corrupted fine-tunes. Through causal interventions, we find that directions isolating social valence, such as the 'Evil' persona vector, and a Semantic Valence Vector (SVV) that we introduce, function as intrinsic guardrails: ablating them drives the misalignment rates above $40$%, while amplifying them suppresses the failure mode to less than $3$%. Leveraging the structural stability of the personality space, we show that vectors extracted $\textit{a priori}$ from an instruct-tuned model transfer zero-shot to successfully regulate EM in corrupted fine-tunes. Overall, our findings suggest that harmful fine-tuning does not overwrite a model's internal representation of personality, allowing conserved representations to serve as robust, cross-distribution guardrails.

Recently, diffusion models have attracted considerable attention for magnetic resonance image reconstruction due to their high sample quality. However, most existing methods rely on large networks with opaque time-conditioning mechanisms, and require offline coil sensitivity estimation. This results in limited interpretability of the reconstruction process and reduced flexibility in the acquisition setup. To address these limitations, we jointly reconstruct the image and the coil sensitivities by combining the parameter-efficient product-of-Gaussian-mixture diffusion model as an image prior with a classical smoothness prior on the coil sensitivities. The proposed method is fast and robust to both contrast and anatomical distribution shifts as well as changing k-space trajectories. Finally, we propose a more expressive parameterization of the image prior which improves results in denoising and magnetic resonance image reconstruction.

Few-shot anomaly detection (FSAD) has made significant strides, yet existing methods still face critical challenges: (i) dependence on task- or dataset-specific training/fine-tuning, (ii) reliance on language supervision or carefully hand-crafted prompts, and (iii) limited robustness across domains. In this paper, we introduce HyperFSAD, a novel FSAD framework that is training-free, language-free, and robust across domains, offering a powerful solution to these challenges. Built upon DINOv3 and a hypergraph-based inference mechanism, our approach performs inference without any task-specific optimization or text prompts, while remaining competitive. Specifically, we replace sensitive nearest-neighbor / top-$n$ matching with \textbf{Sparse Hyper Matching}: \textit{sparsemax} first selects the most relevant support patches, which are then aggregated into a \textit{hyperedge} as compact normal evidence to suppress background noise and distractors. We further introduce \textbf{Dual-Branch Image Scoring}, which fuses \emph{spatial anomaly evidence} from the patch-grid anomaly map with \emph{global semantic deviation} captured by support-aware CLS matching, yielding a robust image-level anomaly score in a strictly visual manner. Notably, all components of HyperFSAD are purely visual, eliminating the need for labor-intensive hand-crafted text prompts. Under the stringent training-free and language-free setting, HyperFSAD achieves state-of-the-art performance across six datasets spanning four industrial datasets (MVTecAD, VisA, MPDD, BTAD) and two medical datasets (RESC, BraTS).

Coreference resolution is typically evaluated using aggregate statistical metrics such as CoNLL-F1, which measure structural overlap between predicted and gold clusters. While widely used, these metrics offer limited diagnostic insights, penalizing errors without revealing whether a system struggles with specific semantic categories, such as people, locations, or events, and making it difficult to interpret model capabilities or derive actionable improvements. We address this gap by introducing a semantically-enhanced evaluation framework for coreference resolution. Our approach overlays Concept and Named Entity Recognition (CNER) onto coreference outputs, assigning semantic labels to nominal mentions and propagating them to entire coreference clusters. This enables the computation of typed scores aimed at evaluating mention extraction and linking capabilities stratified by semantic class. Across our experiments on OntoNotes, LitBank, and PreCo, we show that our framework uncovers systematic weaknesses that remain obscured by aggregate metrics. Furthermore, we demonstrate that these diagnostics can be used to design targeted, low-cost data augmentation strategies, achieving measurable out-of-domain improvements.

Model Predictive Control (MPC) is widely used to operate safety-critical infrastructure by predicting future trajectories and optimizing control actions. However, nonlinear dynamics, hard safety constraints, and numerical optimization often render individual control moves opaque to human operators, undermining trust and hindering deployment. This paper presents Hierarchical Causal Abduction (HCA), which combines (i) physics-informed reasoning via domain knowledge graphs, (ii) optimization evidence from Karush--Kuhn--Tucker (KKT) multipliers, and (iii) temporal causal discovery via the PCMCI algorithm to generate faithful, human-interpretable explanations for control actions computed by nonlinear MPC. Across three diverse control applications (greenhouse climate, building HVAC, chemical process engineering) with expert validation, HCA improves explanation accuracy by 53\% over LIME (0.478 vs. 0.311) using a single set of cross-domain parameters without per-domain tuning; domain-specific KKT-threshold calibration over 2--3 days further increases accuracy to 0.88. Ablation studies confirm that each evidence source is essential, with 32--37\% accuracy degradation when any component is removed, and HCA's ranking-and-validation methodology generalizes beyond MPC to other prediction-based decision systems, including learning-based control and trajectory planning.

Reachability analysis of neural networks, which seeks to compute or bound the set of outputs attainable over a given input domain, is central to certifying safety and robustness in learning-enabled physical systems. Since exact reachable set computation is generally intractable, existing methods typically rely on tractable overapproximations. Examining the state of the art for smooth, twice-differentiable networks, we observe that existing approaches exploit at most second-order information and do not systematically leverage higher-order information. In this work, we introduce \textsc{HiTaB}, a novel verification framework that exploits second-order smoothness through both the Hessian, $\nabla^2 f$, and its Lipschitz constant, $L_{\nabla^2 f}$. We further develop a unified hierarchy of zeroth-, first-, and second-order bounds, together with precise conditions under which higher-order approximations yield provable improvements. Our main technical contribution is a compositional procedure for efficiently bounding $L_{\nabla^2 f}$ in deep neural networks via layerwise propagation of curvature bounds. We extend the framework to both $\ell_2$- and $\ell_\infty$-constrained input sets and show how it can be integrated into branch-and-bound verification pipelines. To our knowledge, this is the first practical reachability analysis framework for smooth neural networks that systematically exploits Lipschitz continuity of curvature, leading to tighter and more informative safety certificates.

Tabular Foundation Models have recently established the state of the art in supervised tabular learning, by leveraging pretraining to learn generalizable representations of numerical and categorical structured data. However, they lack native support for unstructured modalities such as text and image, and rely on frozen, pretrained embeddings to process them. On established Multimodal Tabular Learning benchmarks, we show that tuning the embeddings to the task improves performance. Existing benchmarks, however, often focus on the mere co-occurrence of modalities; this leads to high variance across datasets and masks the benefits of task-specific tuning. To address this gap, we introduce MulTaBench, a benchmark of 40 datasets, split equally between image-tabular and text-tabular tasks. We focus on predictive tasks where the modalities provide complementary predictive signal, and where generic embeddings lose critical information, necessitating Target-Aware Representations that are aligned with the task. Our experimental results demonstrate that the gains from target-aware representation tuning generalize across both text and image modalities, several tabular learners, encoder scales, and embedding dimensions. MulTaBench constitutes the largest image-tabular benchmarking effort to date, spanning high-impact domains such as healthcare and e-commerce. It is designed to enable the research of novel architectures which incorporate joint modeling and target-aware representations, paving the way for the development of novel Multimodal Tabular Foundation Models.

Many studies have shown automatic speech processing (ASR) systems have unequal performance across speakergroups (SG's). However, the manner in which such studies arrive at this conclusion is inconsistent. To pave the wayfor more reliable results in future studies, we lay out best practices for benchmarking ASR fairness based on literaturefrom machine learning fairness, social sciences, and speech science. We first describe the importance of preciselythe fairness hypothesis being interrogated, and tailoring fairness metrics to apply specifically to said hypothesis.We then examine several benchmarks used to rate ASR systems on fairness and discuss how their results can bemisconstrued without assiduous oversight into the intersections between SG's. We find that evaluating fairnessbased on single heterogeneous SG's, such as they are defined in fairness benchmarks, can lead to misidentifyingwhich SG's are actually being mistreated by ASR systems. We advocate for as fine-grained an analysis as possibleof the intersectionality of as many demographic variables as are available in the metadata of fairness corpora in orderto tease out such spurious correlations

Multi-agent LLM systems introduce a security risk in which sensitive information accessed by one agent can propagate through shared context and reappear in downstream outputs, even without explicit adversarial intent. We formalise this phenomenon as propagation amplification, where leakage risk increases across agent boundaries as sensitive content is repeatedly exposed to downstream generators. Existing defences, including prompt-based safeguards, static pattern matching, and LLM-as-judge filtering, are not designed for this setting: they either operate after generation, rely primarily on surface-form patterns, or add substantial latency without modelling the generation process itself. To resolve these issues, we propose PRISM, a real-time defence that treats credential leakage as a sequential risk accumulation problem during generation. At each decoding step, PRISM combines 16 signals spanning lexical, structural, information-theoretic, behavioural, and contextual features into a calibrated risk score, enabling per-token intervention through green, yellow, and red risk zones. Our central observation is that credential reproduction is often preceded by a measurable shift in generation dynamics, characterised by entropy collapse and increasing logit concentration. When combined with text-structural cues such as identifier-pattern detection, these temporal signals provide an early warning of leakage before a secret is fully reconstructed. Across a 2,000-task adversarial benchmark covering 13 attack categories and three pressure levels in a heterogeneous four-agent pipeline, PRISM achieves F1 = 0.832 with precision = 1.000 and recall = 0.712, while producing no observed leakage on our benchmark (0.0% task-level leak rate) and preserving output utility of 0.893. It substantially outperforms the strongest baseline, Span Tagger, which achieves F1 = 0.719 with a 15.0% task-level leak rate.

Large language models (LLMs) can convincingly imitate human writing styles, yet it remains unclear how much stylistic information is encoded in embeddings from any language model and retained after LLM rewriting. We investigate these questions in French, using a controlled literary dataset to quantify the effect of stylistic variation via changes in embedding dispersion. We observe that embeddings reliably capture authorial stylistic features and that these signals persist after rewriting, while also exhibiting LLM-specific patterns. These analytical results offer promising directions for authorship imitation detection in the era of language models.

Expressions with an aspectual variant of a light verb, e.g. 'take on debt' vs. 'have debt', are frequent in texts but often difficult to classify between verbal idioms, light verb constructions or compositional phrases. We investigate the properties of such expressions with a disputed membership and propose a selection of features that determine more satisfactory boundaries between the three categories in this zone, assigning the expressions to one of them.

Designing fair algorithmic decision systems requires balancing model performance with fairness toward affected individuals: More fairness might require sacrificing some performance and vice versa, yet the space of possible trade-offs is still poorly understood. We investigate fairness in binary prediction-based decision problems by conceptualizing decision making as a multi-objective optimization problem that simultaneously considers decision-maker utility and group fairness. We investigate the set of Pareto-optimal decision rules for arbitrary utility functions for decision maker, arbitrary population distributions, and a wide range of group fairness metrics. We find that the Pareto frontier consists of deterministic, group-specific threshold rules applied to individuals' success probability. This complements existing optimality theorems from literature which, for specific fairness constraints, posit lower-bound threshold rules only. However we also show that, depending on the used fairness metric, the Pareto frontier may include upper-bound threshold rules, thus preferring individuals with lower success probabilities. We show that the location of the Pareto frontier depends only on population characteristics, utility functions and fairness score, but not on the technical design of the algorithm - our findings hold for pre-, in-, and post-processing approaches alike. Our results generalize existing optimality theorems for fairness-constrained classification and extend them to generalized fairness metrics and fairness principles, and to partial fairness regimes. This paper connects formal fairness research with legal and ethical requirements to search for less discriminatory alternatives, offering a principled foundation for evaluating and comparing algorithmic decision systems.

AI deployment in sensitive domains such as health care, credit, employment, and criminal justice is often treated as unsafe to authorize until model internals can be explained. This often leads to an excessive reliance on mechanistic interpretability to address a deployment challenge beyond its intended scope. We argue that the gate should instead be calibrated verification: authorization should be domain-scoped, independently checkable, monitored after release, accountable, contestable, and revocable. The reason is twofold. First, model capability is uneven across nearby tasks, so authorization must attach to a specific use rather than to a model in general. Second, societies have long governed opaque expertise through credentials, monitoring, liability, appeal, and revocation rather than mechanism-level explanation. Recent evidence reinforces this distinction between mechanistic understanding and deployment authority: a 53-percentage-point gap between internal representations and output correction shows that understanding may not translate into action, while one scoping review found that only 9.0% of FDA-approved AI/ML device documents contained a prospective post-market surveillance study. We propose Verification Coverage, a six-component reportable standard with a minimum-composition rule, as the metric that should sit beside capability scores in model cards, leaderboards, and regulatory disclosures.