Papers

Spectral computed tomography (CT) extends conventional CT by measuring attenuation across multiple energy channels, allowing improved modeling of physical X-ray interactions and energy-dependent material behavior and leading to richer scene understanding. We present a novel method for spectral CT reconstruction and basis material decomposition using 3D Gaussian Splatting by adding per-Gaussian basis material fractions to the set of learnable parameters, which together with a set of energy-dependent basis functions define the attenuation across the full spectral range. The basis functions represent various physical attenuation models such as photoelectric absorption and Compton scattering, and are jointly optimized across all energy channels through a differentiable polychromatic forward model, with material decomposition performed via mean-shift clustering of the resulting coefficients. We evaluate our method on a baseline real-world dataset as well as a synthetic dataset that we introduce, comparing against traditional reconstruction algorithms and state-of-the-art learning-based CT reconstruction methods. Our approach outperforms all traditional baselines in novel view synthesis and achieves the best PSNR among all compared methods for spectral CT volume reconstruction, while describing all energy channels with a single shared representation that requires a number of Gaussians comparable to single-channel Gaussian splatting-based CT reconstruction approaches. For basis material decomposition, no traditional or learning-based baseline offers one-step decomposition with direct RGB material segmentation, and our method additionally recovers the photoelectric basis with higher PSNR than traditional pipelines.

Recent topic models leverage pretrained embeddings, but neural architectures produce latent representations without grounding in specific texts, and clustering-based pipelines assign representative documents only post hoc, relying on absolute distances distorted by hubness and anisotropy in high-dimensional spaces. We introduce MARETopic, a training-free framework that casts topic discovery as rank-based prototype selection. After projecting embeddings onto a low-dimensional manifold, MARETopic builds ranked lists encoding ordinal neighborhood structure. A greedy algorithm selects exactly K exemplar documents, real corpus texts, whose neighborhoods cover the corpus. Two variants share this criterion. MARETopic$_\text{Corr}$ scores candidates with a query performance predictor and a rank correlation measure, leading Purity and NMI on the two benchmarks with the most categories, ahead of both neural and clustering-based topic models. MARETopic$_\text{Diff}$ scores them with a rank-based diffusion matrix, needs neither measure, and runs 1.7 to 1.9 times faster. Without a single gradient update, MARETopic leads topic coherence on two of three datasets. A novel inter-topic Maximal Marginal Relevance step raises vocabulary diversity at little cost in coherence. Our code is available at https://github.com/thcastilho/maretopic.

Worst-case Optimal Recovery (OR) and Bayesian learning describe the same Gaussian-quadratic-Hilbert problems in two vocabularies. We sharpen the correspondence - the radius of information equals a nugget-optimized GP posterior variance and is attained by the posterior mean at a closed-form balance nugget - and measure, inside three published Bayesian systems, where the worst-case side pays. The ledger is two-sided: the losses instruct as much as the wins. Morozov calibration tracks a test-access oracle within $1.00$-$1.19\times$ where $\sigma$-blind rules fail, is $4.9$-$6.3\times$ more reproducible across noise draws ($p=0.002$-$0.004$), and is the only deployable rule whose selection survives a change of backend ($1.36\times$ against $12$-$30\times$ for the released weight, ML-II and GCV); tight certificates cover at the information-theoretic floor with no numerical slack. But on exchangeable data split-conformal beats the OR head on interval score, a water-filling prior adds nothing without an oracle noise hint, and under covariate shift the OR band keeps coverage on every dataset yet loses interval score to split-conformal, and to a feature-free constant band, on most cells; what pays is not shift but shift on a learnable target, which a training-free audit statistic predicts before any model is fitted. In Bayesian optimization the certified width is a validity floor whose scalar inflation we prove inert under a checkable margin condition and check at every step. Inertness is graded, not binary, and in the size of the inflation as much as in the objective: $\kappa{=}2$ is inert wherever $\kappa{=}5$ is and on more cells besides, while $\kappa{=}5$ moves half the Ackley seeds and every Griewank seed. Exploration is a shape problem, not a scale one. The design rule: match the guarantee tool to the data regime, and audit the regime first.

End-to-end vision-language models (VLMs) bind visual competence to the scale of their language model: as the language model shrinks, perception and reasoning degrade together. We study an embodied scene question-answering (QA) interface in which vision never enters the language model. A frozen perception stack detects and ranges objects; a deterministic semantic serializer compiles the perceived state, errors included, into decision-aligned text; an unmodified text-only large language model (LLM) answers. On a visible-scope-matched, occlusion-audited campus-robot benchmark, under a prospectively frozen criterion, the serialized interface, using detectors fine-tuned in-domain within each fold, outperforms a zero-shot VLM whose language model has the same 7B scale (0.7892 vs 0.7462), with a larger margin at 3B (0.7673 vs 0.6913). Preregistered decoupling experiments show the gain survives paraphrase, attributing it to decision-aligned computation rather than answer-string leakage, while novel judgment vocabularies bound its scope. The advantage grows as the reader shrinks to 1.5B and reverses at 0.5B, and a ground-truth oracle locates the reader-capability floor. Under matched task supervision the interfaces converge: a VLM fine-tuned with low-rank adaptation (LoRA) overtakes the zero-shot system but only ties an equally supervised text reader (0.8441 vs 0.8396, no statistically resolved difference), and both routes remain perception-bound. Reported perception parameters are comparable to those of the VLM's vision tower, and total compute is not smaller.

Quantum encrypted cloning encodes an unknown state into several encrypted components, each offering an alternative way to recover it later. We show that this redundancy can also serve for one-shot fault diagnosis: instead of inspecting the encrypted state, we measure relational Pauli observables testing consistency conditions imposed by the encoding. The canonical Yamaguchi–Kempf scheme encodes the input into \(n\) signal–key pairs, all carrying the same coherent Bell label, plus a transformed copy of the input qubit. Comparing the labels of two pairs yields a deterministic check that localizes an anomalous pair and identifies its Pauli-error class without revealing the label. Such checks cannot tell whether the signal or the key of that pair is faulty, and no measurement on the pairs can (we prove that they already generate the whole group of deterministic state-blind observables supported there, of rank \(2(n-1)\)). This ambiguity matters: a faulty signal costs one redemption path, a faulty key threatens them all. However, retaining also the transformed input qubit contributes exactly two further independent checks, at any multiplicity, and these suffice to identify any single fault drawn from the single-qubit Pauli set. For three clones this gives six checks in all, and we prove that no smaller set of state-blind observables achieves the same resolution. Across clone multiplicities, parity governs how many components must be measured jointly to reach this resolution, not the resolution itself. These findings are then abstracted into a general framework — state-blind observables, deterministic healthy references, syndrome-induced fault partitions, redemption-oriented sufficiency — yielding relational diagnosability: a syndrome need not identify every fault, only enough of it to select a safe redemption path.

Enterprise analytics agents solve long-horizon tool-use problems over distributed business data, requiring retrieval, reasoning, API calls, code execution, and adaptation to intermediate observations. Supervised fine-tuning (SFT) calibrates tool syntax and teacher-supported behavior, whereas reinforcement learning (RL) can explore reward-supported behaviors beyond demonstrations; applied uniformly, however, RL can perturb already-calibrated skills. We study how to balance SFT and RL under production-mirroring beta APIs. We observe that, in our controlled experiment, checkpoint trajectories retrospectively separated into three regimes: Imitation, where SFT captured reliable teacher behavior; Lift, where both stages helped; and Discovery, where useful reward-observable behavior lay outside reliable teacher support. We leverage this prospectively, using teacher support and reward-observable headroom to route features to SFT only, SFT then RL, increased RL allocation, or further environment development. Across 18 subsequent feature-specific experiments, the diagnostic predicted 15/18 observed trajectories. On GPT-OSS 120B, targeted SFT then RL produced positive point estimates on 7/8 advertiser skills relative to a frontier Control; five positive gains had paired 95% confidence intervals excluding zero, while one skill had a confidence-supported regression. The largest gain was non-disclosure (+11.27 points; 95% CI [+9.72, +12.82]). A separate SME audit surfaced that targeted RL reduces standard leakage from 11.8% to 2.9% and adversarial leakage from 22.9% to 6.8% relative to SFT while preserving actionability (86.2% to 85.7%). In a matched uniform-versus-targeted comparison with shared rewards and optimization, targeted RL improved the seven-skill mean delta from +1.62 to +3.57 while using 43% less incremental RL compute.

Robust Principal Component Analysis (RPCA) is a fundamental technique for extracting low-rank structures from data corrupted by sparse anomalies and noise. While classical RPCA and its stable variants have been widely studied, their computational cost grows prohibitively with data dimensionality. Motivated by emerging trends in quantum machine learning, this paper introduces Quantum-Stable RPCA, the first quantum algorithm for robust low-rank and sparse decomposition under noisy intermediate-scale quantum (NISQ) constraints. Our approach integrates Quantum Singular Value Thresholding (QSVT) for low-rank recovery with Quantum Sparse Approximation (QSA) for anomaly detection, while explicitly modeling gate, decoherence, and measurement errors inherent in quantum hardware. We establish six theoretical results covering recovery guarantees, identifiability, robustness to approximate structure, quantum noise resilience, convergence of alternating minimization, and generalization bounds, thereby extending classical RPCA theory to the quantum regime. Extensive Qiskit-based simulations confirm that Quantum-Stable RPCA delivers significant runtime improvements over classical solvers while maintaining competitive reconstruction accuracy, even under realistic NISQ noise levels. This work provides a blueprint for quantum-enhanced robust learning, bridging machine learning and signal processing paradigms.

Heterogeneous multivariate time-series systems exhibit segment-specific nonlinear dynamics that challenge monolithic forecasting architectures. We propose SegTSim, a big-data-driven segmented temporal simulation framework that integrates segment-specific elasticity modeling with adaptive min-gating, dynamic production relocation optimization, multi-factor data fusion with exchange-rate propagation, and a deep ensemble validation pipeline. The framework is validated on US–Japan automotive trade data from USITC repositories spanning 2015 to 2025, comprising approximately 13000 annual records. Under a 25% perturbation scenario, Japanese import volume declines by 20.4% to 0.93 billion USD, while all output variables maintain coefficients of variation below 3.5% across 1000 ensemble inference runs.

Machine learning is being increasingly used to predict where active wildfires will burn the following day, helping inform evacuation boundaries and containment lines. Most models are evaluated using Average Precision (AP), which summarizes performance across all decision thresholds, although acting on a forecast requires choosing one. We benchmarked five discriminative architectures and one generative model on WildfireSpreadTS using a shared evaluation pipeline and two input configurations. We found that model rankings varied depending on whether performance was measured by AP or by threshold-dependent metrics like F1 and IoU. At the default 0.5 threshold, the highest-AP model flagged 4 to 5 times the observed active-fire area and ranked fifth of six on F1 and IoU, and the most recall-heavy model flagged 16 to 23 times. The three models whose flagged area stayed close to the observed area had AP scores 24 to 37% lower. Across models, we identified three distinct prediction profiles: over-predicting, balanced, and under-predicting, which AP alone could not distinguish. Expanding the input from 7 to 23 channels changed AP by 0.03 on average, against a 0.21 to 0.24 spread across models. Ultimately, these results show that AP alone can favor models whose predictions are poorly suited for operational wildfire forecasting.

Qualitative research is widely used in the human and social sciences, characterized by a deep understanding of phenomena through the interpretation of meanings and contexts. Among qualitative data analysis methods, content analysis stands out as a consolidated technique, which allows for the systematic description and interpretation of textual contents. However, as data volume increases, the time required for organization, reading, and categorization becomes a significant challenge, potentially delaying research development. Therefore, this work aimed to develop a web application to support content analysis, based on Bardin's methodology, targeted at academic researchers. The methodology adopted a mixed approach, combining bibliographic research on content analysis with semi-structured interviews with four experienced researchers, aiming to identify real needs and requirements. Based on these inputs, the Cortex software was developed to assist the researcher in the pre-analysis and material exploration stages, generating suggestions for indices, indicators, and categories with full traceability to original documents. The results demonstrate that Cortex is capable of guiding the researcher through the entire methodological workflow, from corpus configuration to results exportation, acting as a methodological collaborator without replacing the researcher's interpretative autonomy.

Vision-language model (VLM) confidence may change in aggregate when visual evidence is degraded while remaining structurally inconsistent within individual examples. We study answer-level reliability along five-step, question-conditioned evidence-loss trajectories. Using a frozen Qwen2.5-VL-3B-Instruct model, we construct 176 accepted GQA-derived trajectories (880 masking conditions) by progressively masking scene-graph-localized question-critical regions. Native sequence confidence has an evidence monotonicity violation rate (EMVR) of 0.436, and 92.0% of trajectories contain at least one adjacent violation. A matched non-critical-region control shows that full critical masking reduces accuracy by 28.2 percentage points, compared with 0.6 points for equally sized non-critical masks; the paired difference is 27.6 points (95% CI [20.0, 34.7]). We train a lightweight post-hoc reliability head on frozen hidden states, sequence confidence, and entropy. Adding evidence-order supervision to binary cross-entropy (BCE) reduces masking EMVR from 0.330 to 0.303 (paired difference -0.027, 95% CI [-0.044, -0.010]). The same mask-trained objective reduces EMVR from 0.449 to 0.402 on held-out question IDs under unseen local Gaussian blur (difference -0.0468, 95% CI [-0.0739, -0.0199]). AUROC, Brier, and AURC differences between the two learned heads are statistically inconclusive, and native confidence remains stronger for selective-risk ranking. The results separate evidence-order consistency from conventional correctness discrimination rather than establishing generic confidence superiority.

Indirect time-of-flight (iToF) cameras provide compact and cost-effective dense depth measurements, but their ranging accuracy is often degraded by sensor-intrinsic uncertainty under practical imaging conditions. Spatially uniform or range-only Gaussian perturbations cannot accurately reproduce the range-dependent and signal-dependent noise characteristics of real iToF measurements, leading to a synthetic-to-real gap for learning-based restoration. To address this problem, we propose a joint depth-uncertainty modeling and restoration framework for reliable iToF sensing. A sensor-intrinsic depth-uncertainty model is first developed from calibrated tap responses, returned-signal levels, and sensor noise statistics through a depth-oriented weighted least-squares formulation. The resulting pixel-wise uncertainty is used for heteroscedastic depth synthesis and uncertainty-aware restoration supervision. Based on this heteroscedastic data synthesis, we further develop a U-shaped restoration network with Depth Visual State Space (DVSS) blocks, which combine long-range state-space modeling with convolutional spatial-channel refinement for structure-preserving depth recovery. Experiments on synthetic data and measurements captured by an in-house iToF prototype validate the proposed uncertainty model under varying range and returned-signal conditions. Controlled comparisons with fixed and range-aware Gaussian noise, together with evaluations on U-Net, Restormer, and DVSS, further demonstrate that the proposed synthesis consistently benefits different restoration backbones. The complete framework achieves 40.85~dB PSNR and 2.54 mm MAE on the synthetic test set, and 35.42 dB PSNR and 4.87 mm MAE on real iToF measurements.

We propose ECHO for dyadic 3D facial motion generation under a strict dual-stream audio-only setting, formulating the problem as an asymmetric task involving speech-constrained articulation and one-to-many listener reactions. To address this asymmetry, ECHO decomposes motion into a deterministic anchor that captures stable speech-correlated structure and a stochastic residual that models the remaining one-to-many interaction dynamics. On top of this backbone, Motion Memory acts as a training-only regularizer during brief late-stage fine-tuning to provide local priors for weakly conditioned listening windows, while semantic-group scaling controls residual injection across expression, jaw, and neck. This design balances speaking-side articulatory fidelity with listening-side realism and diversity in a single generation process. Results from unified, state-wise, and ablation evaluations show that conversational 3D motion benefits from decomposing stable and uncertain components rather than applying stochasticity uniformly. ECHO provides a practical formulation and technical basis for deployable conversational digital humans under strict audio-only conditions.

Systematic reviews require sustained human judgment across thousands of records, yet existing evaluations of large language models (LLMs) typically examine review stages in isolation. We introduce SciLitBench, a multi-stage benchmark spanning title and abstract screening, full-text screening, and schema-guided data extraction, with 42,981 retrieved records, 1,012 full texts, and annotations for 888 included papers. Across 22 open-weight LLMs from six model families, explicit inclusion and exclusion criteria improve title and abstract screening $F_2$ by 28.8%, while researcher-authored rationales improve full-text screening by 15%. Data extraction reveals a different reliability regime: performance declines from 0.97 accuracy for publication year to 0.37 Jaccard overlap for computational approach, while the strongest models recover only 30% of annotated evaluation evidence and 25% of limitations. SciLitBench identifies a practical boundary between high-recall screening and evidence-complete extraction and provides a reproducible resource for evaluating LLM-assisted evidence synthesis.

The creation of telescope bibliographies is a crucial part of assessing the scientific impact of observatories and ensuring reproducibility in astronomy. This task involves identifying, categorizing, and linking scientific publications that reference or use specific telescopes. However, this process remains largely manual and resource intensive. In this work, we present an efficient SciBERT-based approach for automatic classification of scientific papers into four categories - science, instrumentation, mention, and not telescope. Despite strict context-length constraints (maximum 512 tokens) and limited compute resources, our approach achieved a macro F1 score of 0.89, ranking at the top of the WASP-2025 leaderboard. We analyze the effect of truncation and show that even with half the samples exceeding the token limit, SciBERT's domain alignment enables robust classification. We discuss trade-offs between truncation, chunking, and long-context models, providing insights into the efficiency frontier for scientific text curation.

The question of what constitutes a system remains fundamental to General Systems Theory. Existing definitions commonly characterize a system in terms of elements, relationships, boundaries, functions, or interactions, but these perspectives do not always provide a unified account of how system structure and system behavior constitute one another. This paper proposes Structure-Behavior Coalescence (SBC) as an interaction-based account of what a system is. From the SBC perspective, a system is not merely a collection of elements or relationships, nor is it adequately characterized by behavior considered independently of structure. Rather, a system is a structured entity whose behavior arises through interactions among its constituent entities, with those interactions simultaneously contributing to both its structural organization and behavioral realization. The paper develops this perspective by distinguishing system structure, interaction, and behavior while treating them as inseparable aspects of a unified system representation. It argues that interactions provide the essential link through which structural relationships become behavioral processes and through which behavioral processes reveal and instantiate system structure. This perspective provides a basis for representing systems in a manner that maintains consistency between what a system is and what a system does. The paper further considers the implications of SBC for General Systems Theory, particularly for system identity, system boundaries, behavioral emergence, and the representation of complex systems. SBC is presented as a general conceptual foundation that complements existing systems theories by placing the coalescence of structure and behavior through interaction at the center of the definition of a system.

This paper examines a foundational assumption in traditional systems theory, namely that structure (the organization of components) and behavior (the evolution of system activity over time) can be treated as separable analytical dimensions. It argues that this separation contributes to persistent difficulties in explaining system identity, particularly in cases involving change, emergence, and boundary specification. To address this issue, the paper introduces Structure-Behavior Coalescence (SBC) as a reframing principle. SBC proposes that structure and behavior should not be understood as independently existing entities that are subsequently related through modeling constructs, but as mutually constitutive aspects of a single systemic process. From this perspective, system identity is understood as arising from the sustained co-determination of structural organization and behavioral dynamics, rather than from their external correspondence or alignment. This reframing provides a unified way of understanding system identity, emergence, and boundary formation within a cybernetically informed systems perspective.

A direct test of an LLM judge of agent trajectories injects faults into correct runs and reports recall, per fault type or by whether the fault broke the environment outcome (loud) or not (silent). Such recall can credit a judge with detection it does not have; paired discrimination, its flag rate on the faults minus its rate on the clean runs they came from, exposes this. Our testbed, a deterministic support desk with a scripted oracle and a one-step fault injector, labels all 400 trajectories exactly. A 14B judge shown only the request and final reply scores 34% to 76% recall on four fault types that leave the reply unchanged. There its input is the clean run's, so its paired discrimination is zero and that recall is its flag rate on clean runs. Splitting by outcome survival does not fix this: its loud recall of 84% is a paired +0.393 and its silent recall of 45% a paired +0.048, all from the two fault types that change the reply. Told to check each step, the same model flags every fault of those four types and 0 of 100 clean runs (95% CI up to 3.6%). It does not reliably check the reply: of four invented promises it flags one every time and the other three once in 42 faults. Shown every step but asked only about the reply, it still reaches a paired +0.69 on reply-unchanged faults, against +1.00 when told to check each step. We recommend reporting paired discrimination against clean parents, split by whether the fault reaches the judge's input and by outcome survival, and release the testbed, raw verdicts and analysis pipeline.

The total scaled-gradient variation (TSGV) regularizer, derived from sparse modeling of piecewise-linear structures, has been shown to preserve edges and corners in image restoration. However, its highly nonconvex and nonlinear nature poses severe computational challenges, as existing methods often suffer from parameter sensitivity or lack convergence guarantees. To overcome this, we propose a tailored bilinear decomposition that decouples the nonlinear weighted gradient in the TSGV regularizer. This approach yields an equivalent optimization problem governed by cone or sphere constraints, depending on the chosen scaling function. In particular, the cone constraint plays a central role in characterizing edge- and corner-preserving behavior. We solve this reformulation using the alternating minimization method (AMM) equipped with a majorization–minimization strategy, ensuring a monotonic decrease in energy without step-size tuning. Furthermore, we provide a geometric interpretation of the edge-preserving properties of these constraints by analyzing their asymptotic behavior near image singularities. We establish the global convergence of the proposed method to a critical point within the Kurdyka–Łojasiewicz framework. Extensive numerical experiments on Gaussian denoising and non-line-of-sight (NLOS) imaging show that the proposed method achieves PSNR and SSIM competitive with or superior to representative variational methods, especially at high noise levels, and improves the structural reconstruction under dense and sparse scanning.

An LLM agent calling a production API cannot distinguish a query that matched nothing from a query the server did not understand. Both return HTTP 200 with a parsable body, no exception to catch and no field to branch on. We ask what predicts which one occurred, and what it does to the agent. Auditing 721,320 parameters across 2,501 independently published OpenAPI documents, we find that 7.5% declare an enumeration and 15.2% declare any machine-checkable constraint at all, while 40.1% of documents state at least one constraint in prose that their schema does not encode. Executing 219 schema-derived perturbations against live commercial endpoints from 27 vendors, reached through a single aggregation layer (Monid) that publishes a schema and returns a run identifier for every call, we find that constraint form, not vendor identity, predicts honesty: machine-checkable constraints yielded an honest error in 111 of 111 cases, prose-only constraints failed silently in 44 of 61 (p = 2e-13). Twelve models across eight families then met these endpoints on ordinary tasks. A vocabulary that the description merely exemplifies was missed by every model on 88 of 88 attempts, while vocabularies written out in full were used correctly 88 to 91% of the time. Running the full agent loop, models detected the resulting silent failure in 12% of cases, repaired it in 0%, asserted a false negative to the user in 41%, and invented a figure in 12%. Promoting the vocabulary into the schema removes the failure, from 88 of 88 to 0 of 89. The fix is one line of schema rather than a better model. Code, schemas, perturbation sets, agent transcripts and per-call run identifiers are released at https://github.com/Jasper0122/silentprobe.

Knowledge distillation trains a small student model to reproduce the outputs of a large teacher model, and its progress is typically monitored through the teacher–student discrepancy. The quantity of ultimate interest, however, is the student's error with respect to the true task. We study the relation between these two objectives in a minimal three-party model, a true teacher (generative model), a teacher, and a student, all soft committee machines, in which the true teacher contains a shared latent factor that the teacher cannot represent, with mismatch strength controlled by a single scalar $\dmiss$. Within an order-parameter description of online distillation, and exploiting closed-form (arcsine-type) expressions for all errors under error-function activations, we prove that the learning dynamics and the distillation error $\Ets$ are exactly invariant to $\dmiss$, whereas the true error $\Etzs$ and the gap $\Delta=\Etzs-\Ets$ are strictly increasing in $\dmiss$, with a rate that is amplified linearly by the complexity $M_0$ of the true teacher. Numerical phase diagrams over the plane spanned by true-teacher complexity and student capacity confirm the predicted deformation: the contours of $\Ets$ do not move while the landscape of $\Etzs$ rises systematically, and a teacher-miss regime, where mimicry succeeds but the task fails, expands with $\dmiss$. The results give a quantitative warning against evaluating distillation solely through teacher-mimicry metrics and identify the gap $\Delta$ as a minimal diagnostic for distinguishing teacher-miss from capacity-limited failure.

Zero-shot anomaly detection (ZSAD) has gained significant attention for its practical value in industrial inspection. Recently, CLIP-based approaches have been widely adopted in ZSAD due to their strong vision-language generalization capabilities. However, existing methods commonly employ continuous prompt embeddings for prompt optimization and encode semantics in latent vectors, which lack interpretability and scalability. To this end, we propose CoEvoAD, a co-evolutionary framework for discrete prompt selection. CoEvoAD performs prompt search in the discrete natural-language space using an evolutionary algorithm. Candidate prompts are iteratively generated, evaluated, and selected throughout population evolution, thus preserving the interpretability and composability of natural language. Furthermore, we introduce a Cross-Category Transfer Objective (CCTO), which treats held-out source categories as proxies for unseen categories and scores prompt rules based on their estimated cross-category transferability, effectively improving cross-category generalization. Extensive experiments are conducted to validate the effectiveness of CoEvoAD, and the results show that it achieves state-of-the-art performance across multiple anomaly detection datasets. The code is available at https://github.com/rstao-bjtu/CoEvoAD.

Deployed decisions are often optimized once and retained because updates impose operational, regulatory, or switching costs. As operating conditions change, when should such decisions be re-optimized? We study this question for stochastic optimization when the objective's functional form is known but the decision maker's trade-offs are encoded by an unknown preference parameter. Standard distribution-shift tests are poorly aligned with this goal: they can flag detectable yet decision-irrelevant changes without determining whether the incumbent decision has become materially suboptimal. We propose RADAR (Regret-based Assessment of Decision Adequacy and Risk), a decision-focused framework that uses inverse optimization to infer latent preferences and tests the deployed decision's optimality gap under the current distribution. By targeting regret, RADAR ignores decision-irrelevant shifts while detecting changes that warrant re-optimization. We develop two-sample and sequential changepoint procedures and establish asymptotic guarantees for Type-I error and power. Across synthetic optimization problems, a semi-synthetic capacity allocation task, and police-zone planning, RADAR more reliably distinguishes harmful from harmless shifts than decision-agnostic alternatives.

Chain-of-thought (CoT) monitoring assumes that reasoning traces faithfully record the information that shapes a model's answer. Existing faithfulness tests often place explicit bias cues in the user message, while agents may encounter preferences through tool returns or raw artifacts. We introduce FACE-Eval (Faithful Attribution of Cue Effects Evaluation), a 5,100-sample evaluation that varies cue location (user message or tool return) and explicitness (direct summary or raw artifact). We measure verbalized commitment among cue-following answers and unverbalized adoption among all cued samples. We evaluate 15 open-weight models from eight families, with total parameters ranging from 4B to 1.60T. Every model has lower verbalized commitment for tool-return than user-message cues and for implicit than explicit cues. Unverbalized adoption is higher for tool-return cues on all 15 models and for implicit cues in 28 of 30 model-channel comparisons. A source-attribution prompt narrows the channel gap on seven models, sometimes by increasing user-channel unverbalized adoption, while telling models that their reasoning will be monitored does not reliably close the gap. We also use two transcript monitors (GPT-5.6-Luna and GPT-4o-mini) to detect preference adoption in the largest model of each family. Across 32 model-channel-explicitness cells, higher unverbalized adoption is associated with lower detection ability for both monitors (Pearson r=-0.54 and r=-0.78, respectively). These results suggest that CoT monitoring may be less reliable when preference information arrives through tools or must be inferred from raw artifacts, within the single-call, prefilled-tool setting tested here.

A benchmark score is a joint property of the model, the evaluation harness, the elicitation budget, the sampled population, and contamination status. Leaderboards publish the model and the score, so capability and leakage stay observationally equivalent. Existing taxonomies classify contamination for automated detection, not the question a reporter faces at publication: given the mitigations already applied, which validity threats remain open? We introduce a taxonomy organized by the mitigation each type defeats – direct, derivative, temporal, distributional, and acquired – spanning training-time and evaluation-time leakage. Holding out a private test set closes the first alone. The fifth is acquired during the evaluation itself; because it is a property of one run, it must be recorded with the reported score rather than with the benchmark release. We operationalize it as a four-field disclosure protocol in which "unknown" is a valid entry, released under CC BY 4.0 with a JSON Schema, a validator, and worked examples. Two coders external to the design team applied a pre-registered instrument to 41 documents. Per-variable linear-weighted $\kappa$ runs from 0.00 to 0.35 (median 0.21) over 29 main-pass documents against a single-coder test-retest ceiling of 0.84, collapsing under the class skew the registration anticipated; pooling raises it to 0.46 through chance correction rather than better agreement. Two variables fall below the prevalence-robust threshold registered in advance: strata reporting and the acquired type introduced here. Disagreement concentrates on when a variable applies rather than on what a document states. Elicitation budgets are reported in 13% of documents, and no document addresses all five types. The contribution is the taxonomy, the score-side artifact that follows from it, and a pre-registered measurement of instrument reliability and current disclosure.

Sandbagging models strategically underperform on evaluations while retaining the capabilities being measured. The evaluations that guide frontier-model deployment and governance then understate what these models can do. To understand the mechanism, we propose a causal model of how sandbagging is carried in the residual stream. Early layers write the sandbagging intent onto a single axis of the stream, and a later layer reads that axis and commits the answer. We study three instruction-tuned models (Qwen2.5-7B, Llama-3-8B, and Mistral-7B) and four ways of installing a sandbagging lock (prompting, fine-tuning, reinforcement learning, and circuit breaking). Each lock decides from the prompt whether to sandbag, and the fine-tuned and circuit-broken locks answer honestly whenever a password appears. The causal model predicts a window of layers, after the last sandbagging write and before the answer commit, in which a single-layer reference graft of the sandbagging axis to its honest value restores the full capability. The single-layer graft recovers the capability in 28 of the 33 runs of the prompted, fine-tuned, and RL-trained locks, with a median held-out recovery of 96%. The circuit-broken lock rewrites the whole state through a band of layers, and the single-layer graft fails at every layer. We therefore introduce a second intervention, context grafting, which replays the password's cached key/value activations so that every layer's attention reads them as additional context. Context grafting provably and empirically restores the full capability on all three models, and the recovery is surprisingly insensitive to the exact password content. More broadly, an auditor can use this causal model to design interventional auditing techniques for sandbagging models.

When coding agents encounter defective test infrastructure they may reward-hack: hardcoding outputs or editing test files to pass tests they cannot legitimately satisfy, a pattern that has now appeared outside benchmarks, in a coordinated multi-agent intrusion of a major AI platform's production infrastructure. The same capability that lets an agent detect and exploit a defect could let it report one, given the right decision environment. We evaluate escalation channels, structured reporting tools available to the agent at the point of conflict, as a decision-environment intervention that both reduces reward hacking and surfaces the infrastructure defects that trigger it. A $2 \times 2$ factorial separates the contributions of an escalation tool, a standalone anti-reward-hacking policy, and their combination. Across 8 frontier models spanning 5 families, the combined intervention reduces reward hacking from 23.6% to 5.3% (mixed-effects logistic OR = 9.2, 95% CI 5.0–16.8, $p < 10^{-12}$) with no detectable cost or performance overhead, eliminating it entirely for 6 of 8 models. Escalation and hacking are near-perfectly mutually exclusive, with 98.7% of escalations involving no hacking (100% under the combined intervention). Beyond reduction, escalation channels function as diagnostic infrastructure: on top of monitoring, escalation adds +10.1 percentage points of defect detection coverage and is more accurate once it fires (99.4% vs 85.8%). Unlike containment-based approaches that risk outpacing growing model capabilities, escalation channels redirect capability toward disclosure rather than exploitation.

Skills, as a useful abstraction for the procedural capabilities of large language models (LLMs), capture how models perform structured, multi-step reasoning and program execution. Existing approaches typically treat skills as explicit, surface-level constructs specified through prompts or programs, leaving open the question of how such procedural capabilities are represented inside the model and whether they can be manipulated as structured objects in latent space. In this empirical study, we investigate whether procedural LLM skills can be represented as directions in activation space and whether vector-space operations over these directions can express skill-level behaviors. We find that procedural skills admit a vector-space representation: individual skill directions can be activated to shift model behavior; independently extracted directions can compose to form higher-level skills. Contrastive directions yield context-conditioned algorithmic personalization and optimization trajectories over skill directions evolve non-monotonically, with intermediate states often surpassing fully optimized solutions. These results support a representation-level view of procedural LLM skills: they admit a latent vector-space organization that allows direct manipulation through internal interventions.

Sandbagging, in which a model deliberately underperforms on an evaluation despite retaining the underlying capability, threatens the safety evaluations that frontier-model governance depends on. The Elicitation Game found that fine-tuning elicits hidden capability from sandbagging model organisms whereas additive activation steering fails. We revisit that verdict with reference-grafting, which sets an activation's coordinate along a contrast direction to the value it takes in an honest reference, at a small set of circuits chosen by active learning. Across eleven fine-tuned password-locked models (1.5-32B, three architecture lineages), it recovers +94 to +101% of the honest-sandbagging gap, matching fine-tuning elicitation without weight updates or training labels; two to five paired examples suffice to fit the direction. Similar recovery holds for reinforcement-learning-induced sandbagging and for password-locked code generation. Grafting works because the fine-tuned lock is a thresholded gate: held-out accuracy stays at the sandbagged level until the grafted coordinate crosses a threshold near the honest reference, which is why additive steering and zeroing the coordinate often fail. The direction tracks how the sandbagging was induced rather than what is withheld – fit on grade-school science (ARC) it elicits withheld hazardous knowledge (WMDP), yet prompting, fine-tuning, and reinforcement learning each carry a different direction. Circuit-breaking marks the boundary: it reroutes activations on every forward pass, so the fixed edits we test are re-broken downstream and do not restore coherent generation.

We analyse a non-residual variational multiscale finite element formulation with dynamic subscales for incompressible generalised Newtonian Navier–Stokes flows. The apparent viscosity is assumed to be bounded and Lipschitz continuous with respect to the shear rate, covering several regularised rheological laws. The method separates, through orthogonal projections, the unresolved pressure-divergence, convective, and pressure-gradient contributions. For the linearised semi-discrete problem, we prove well-posedness and an unconditional stability estimate in an anisotropic VMS norm. The estimate controls viscous dissipation, subscale energies, discrete divergence, and a coupled acceleration-convection-pressure balance. For the non-linear formulation, a fixed-point argument yields existence and optimal-order a priori bounds under suitable regularity and smallness assumptions. The analysis also provides weak-in-time control of the discrete pressure-convection balance. A key point is that the dynamic pressure-related subscale provides the time-derivative contribution needed to control the projected pressure-gradient error, a mechanism unavailable in the corresponding quasi-static setting.