Large language models (LLMs) are increasingly trusted as automated judges, assisting evaluation and providing reward signals for training other models, particularly in reference-based settings like Reinforcement Learning with Verifiable Rewards (RLVR). However, we uncover a critical vulnerability even in this reference-based paradigm: generative reward models are systematically susceptible to reward hacking. We find that superficial inputs, which we term ''master keys'' such as non-word symbols (e.g., '':'' or ''.'') or generic reasoning openers (e.g., ''Thought process:'' or ''Let's solve this problem step by step.''), can consistently elicit false positive rewards without any substantive reasoning. Our systematic evaluation demonstrates this is a widespread failure affecting a diverse range of models, including leading proprietary systems such as GPT-o1 and Claude-4. These results challenge the assumed robustness of LLM judges and pose a significant threat to their reliability. To address this, we propose a simple yet effective data augmentation strategy using truncated model outputs as adversarial negative examples. The resulting Master Reward Models (Master-RMs) demonstrate state-of-the-art robustness against these ''master key'' attacks while maintaining high performance in standard evaluation settings. We supplement these findings with a comprehensive analysis of the vulnerability across model scales, prompt variations, and common inference-time strategies, offering insights to guide future research on robust LLM evaluation. We release our robust, general-domain reward models and the synthetic training data at https://huggingface.co/sarosavo/Master-RM and https://huggingface.co/datasets/sarosavo/Master-RM.

Deep learning-based machine listening is broadening the scope of industrial acoustic analysis, yet its widespread implementation on live shop floors is hindered by the reliance on large, task-specific annotated datasets for every new task. While emerging general-purpose sound foundation models aim to alleviate data dependency, they reveal critical dilemmas in practice. General-purpose sound foundation models are computationally expensive and fail in industrial scenarios characterized by tonal harmonics, broadband noise, and transient fault events, making instant, on-site deployment impractical. These challenges combined mean that a practical, end-to-end system for deploying a sound foundation model on a live shop floor has remained elusive. To address this challenge, this study introduces LISTEN (Lightweight Industrial Sound-representable Transformer for Edge Notification), the first lightweight foundation model specialized for industrial sound. Through Knowledge Distillation (KD) from the large-scale teacher model IMPACT (Industrial Machine Perception via Acoustic Cognitive Transformer), we construct LISTEN optimized for resource-constrained edge environments. By freezing the backbone and training only a shallow head on minimal target-process data, rather than performing full fine-tuning or retraining, LISTEN achieves nearly identical performance to IMPACT across diverse manufacturing processes. This study further demonstrates a complete system for real-time machine monitoring, encompassing data acquisition with Industrial Internet of Things (IIoT) devices, rapid model adaptation using minimal annotated data, and real-time monitoring on a low-cost edge device. By validating the entire system on a live CNC machine, this work establishes the first feasible end-to-end system for deploying a lightweight industrial sound foundation model in an active industrial environment.

In this paper we analyze the behaviour of the stochastic gradient descent (SGD), a widely used method in supervised learning for optimizing neural network weights via a minimization of non-convex loss functions. Since the pioneering work of E, Li and Tai (2017), the underlying structure of such processes can be understood via parabolic PDEs of Fokker-Planck type, which are at the core of our analysis. Even if Fokker-Planck equations have a long history and a extensive literature, almost nothing is known when the potential is non-convex or when the diffusion matrix is degenerate, and this is the main difficulty that we face in our analysis. We identify two different regimes: in the initial phase of SGD, the loss function drives the weights to concentrate around the nearest local minimum. We refer to this phase as the drift regime and we provide quantitative estimates on this concentration phenomenon. Next, we introduce the diffusion regime, where stochastic fluctuations help the learning process to escape suboptimal local minima. We analyze the Mean Exit Time (MET) and prove upper and lower bounds of the MET. Finally, we address the asymptotic convergence of SGD, for a non-convex cost function and a degenerate diffusion matrix, that do not allow to use the standard approaches, and require new techniques. For this purpose, we exploit two different methods: duality and entropy methods. We provide new results about the dynamics and effectiveness of SGD, offering a deep connection between stochastic optimization and PDE theory, and some answers and insights to basic questions in the Machine Learning processes: How long does SGD take to escape from a bad minimum? Do neural network parameters converge using SGD? How do parameters evolve in the first stage of training with SGD?

We prove real-rootedness for the Poincaré polynomial \[ P_n(t)=\sum_{i=0}^{n-3} \dim H^{2i}(\overline{\mathcal M}_{0,n};\mathbb{Q})t^i \] of the Deligne--Mumford moduli space $\overline{\mathcal M}_{0,n}$ of stable $n$-pointed rational curves, proving a conjecture of Aluffi--Chen--Marcolli. The proof starts from the Keel--Manin--Getzler recurrence, but its main new idea is a bivariate deformation $F_m(y,t)$ of the Poincaré polynomial. This deformation reveals a hidden interlacing structure not visible in the one-variable recurrence. For fixed $t<0$, the zero set of $F_m$ in the $y$-direction is controlled by a Sturm--Rolle argument on the interval $0<y<1-t$. The original polynomial is recovered on the slice $y=1$, and the ordered crossings of the moving roots through this slice give both real-rootedness and strict interlacing. Consequently, the Betti numbers of $\overline{\mathcal M}_{0,n}$ form an ultra-log-concave sequence. We further prove real-rootedness and ultra-log-concavity for the Poincaré polynomial of the Fulton--MacPherson space $\mathbb{P}^1[n]$ of $n$ ordered points in degenerations of the complex projective line. The proof for $\overline{\mathcal M}_{0,n}$ was obtained through an iterative AI-assisted workflow with Co-Mathematician, an agentic frontier-model system developed by Google DeepMind. Our role was to formulate the problem, evaluate the proposed proof attempts, identify gaps and request corrections, compare the developing argument with the literature, and refine the presentation of the final proof. Our additional human contribution was to observe that a similar residual deformation strategy applies to the Fulton--MacPherson spaces $\mathbb P^1[n]$, yielding the corresponding real-rootedness theorem.

Spracklen et al. (USENIX Security '25) showed that code-generating large language models hallucinate package names that do not exist on PyPI or npm at rates ranging from 5.2% on commercial models to 21.7% on open-source models, creating an attack surface for slopsquatting -- the registration of malicious packages under hallucinated names. We replicate their methodology on five frontier code-capable LLMs released between October 2025 and March 2026: Claude Sonnet 4.6, Claude Haiku 4.5, GPT-5.4-mini, Gemini 2.5 Pro, and DeepSeek V3.2. Across 199,845 paired Python and JavaScript prompts validated against PyPI and npm master lists, we measure overall hallucination rates between 4.62% (Claude Haiku 4.5) and 6.10% (GPT-5.4-mini) -- an order-of-magnitude compression of the inter-model spread observed by Spracklen, but not a retirement of the threat. Beyond replication, we identify a set of 127 package names (109 on PyPI, 18 on npm) that all five evaluated models invent identically; following coordinated disclosure with PyPI Security and Socket.dev, 53 of these (41 on PyPI, 12 on npm) remain registrable by an attacker after each registry's existing defenses, constituting a model-agnostic supply-chain attack surface that no single-model study can reveal. We further document a Python-over-JavaScript hallucination asymmetry that inverts Spracklen's 2024 finding, identify a Haiku-below-Sonnet inversion within the Anthropic family, and observe a Jaccard-similarity peak between DeepSeek V3.2 and GPT-5.4-mini (J = 0.343) suggestive of shared training-data origins.

Precision oncology is currently limited by the small-N, large-P paradox, where high-dimensional genomic data is abundant but pharmacological response samples are sparse. While deep learning achieves predictive accuracy, it frequently fails to provide the mechanistic clarity required for clinical adoption. We present the Contextual Invertible World Model (CIWM), a Neuro-Symbolic Agentic Framework that bridges this gap by integrating a quantitative machine learning emulator with a Large Language Model reasoning layer. Utilising a stringently curated, high-fidelity data engineering pipeline on the Sanger GDSC dataset (\( N=83 \)), we isolate true biological signals from in vitro artifacts to establish a rigorous baseline predictive correlation for complex transcriptomics (\( r=0.268 \)). Through Inverse Reasoning, we perform in silico CRISPR perturbations across the colorectal landscape. The framework autonomously overturns classical mechanistic assumptions, identifying a hierarchical dominance of mutant KRAS over the APC/Wnt-axis in driving 5-fluorouracil resistance (\( Δ=-0.0469 \)) via a "KRAS Shield" mapped to MAPK/PI3K networks. Furthermore, the agentic layer identified a "PIK3CA Paradox", revealing that repairing PIK3CA inadvertently increases chemoresistance (\( Δ=+0.0085 \)) by triggering a compensatory feedback loop that hyperactivates the dominant MAPK survival pathway.

Dynamic functional connectivity (dFC) derived from resting-state functional magnetic resonance imaging (fMRI) has been extensively utilized in brain science research. The sliding window correlation (SWC) method is a widely used approach for constructing dFC by computing correlation coefficients between amplitude time series of signals from pairs of brain regions. In this study, we propose an integrated approach that incorporates both amplitude and phase information of fMRI signals to improve the detection of brain disorders. Specifically, we introduce a multi-scale fusion learning framework, namely MSFL, which leverages two complementary dFC features derived from SWC and phase synchronization (PS). Here, SWC captures amplitude correlations, while PS measures phase coherence within dFC. We evaluated the efficacy of MSFL in classifying autism spectrum disorder and major depressive disorder using two publicly available datasets: ABIDE I and REST-meta-MDD, respectively. The results indicate that MSFL significantly outperforms existing comparative models. Moreover, we performed model explanation analysis using the SHAP framework, which showed that both types of dFC features from SWC and PS contribute to detecting brain disorders.

Machine learning models have recently emerged to predict whether hypothetical solid-state materials can be synthesized. These models aim to circumvent direct first-principles modeling of solid-state phase transformations, instead learning from large databases of successfully synthesized materials. Here, we assess the alignment of several recently introduced synthesis prediction models with material and reaction thermodynamics, quantified by the energy with respect to the convex hull and a metric accounting for thermodynamic selectivity of enumerated synthesis reactions. A dataset of successful synthesis recipes was used to determine the likely bounds on both quantities beyond which materials can be deemed unlikely to be synthesized. With these bounds as context, thermodynamic quantities were computed using the CHGNet foundation potential for thousands of new hypothetical materials generated using the Chemeleon generative model. Four recently published machine learning models for synthesizability prediction were applied to this same dataset, and the resultant predictions were considered against computed thermodynamics. We find these models generally overpredict the likelihood of synthesis, but some model scores do trend with thermodynamic heuristics, assigning lower scores to materials that are less stable or do not have an available synthesis recipe that is calculated to be thermodynamically selective. In total, this work identifies existing gaps in machine learning models for materials synthesis and introduces a new approach to assess their quality in the absence of extensive negative examples (failed syntheses).

In this paper, we demonstrate how to enhance the validity of causal inference with unstructured high-dimensional treatments like texts, by leveraging the power of generative Artificial Intelligence (GenAI). Specifically, we propose to use a deep generative model such as large language models (LLMs) to efficiently generate treatments and use their internal representation for subsequent causal effect estimation. We show that the knowledge of this true internal representation helps disentangle the treatment features of interest, such as specific sentiments and certain topics, from other possibly unknown confounding features. Unlike existing methods, the proposed GenAI-Powered Inference (GPI) methodology eliminates the need to learn causal representation from the data, and hence produces more accurate and efficient estimates. We formally establish the conditions required for the nonparametric identification of the average treatment effect, propose an estimation strategy that avoids the violation of the overlap assumption, and derive the asymptotic properties of the proposed estimator through the application of double machine learning. Finally, using an instrumental variables approach, we extend the proposed GPI methodology to the settings in which the treatment feature is based on human perception. The GPI is also applicable to text reuse where an LLM is used to regenerate existing texts. We conduct simulation and empirical studies, using the generated text data from an open-source LLM, Llama 3, to illustrate the advantages of our estimator over state-of-the-art causal representation learning algorithms.

We propose KG-ER, a conceptual schema language for knowledge graphs that describes the structure of knowledge graphs independently of their representation (relational databases, property graphs, RDF) while helping to capture the semantics of the information stored in a knowledge graph.

An adaptive bandwidth selection procedure for the mixture kernel in the maximum mean discrepancy (MMD) for fitting generative moment matching networks (GMMNs) is introduced, and improved learning of copula random number generators is demonstrated. Based on the relative error of the training loss, the number of kernels is increased during training; additionally, the relative error of the validation loss is used as an early stopping criterion. While training time remains similar, adaptively training GMMNs (AGMMNs) significantly increases training performance, which is shown based on validation MMD trajectories, samples and validation MMD values. Superiority of AGMMNs over GMMNs and parametric copula models is also demonstrated in terms of three applications. First, convergence rates of estimators based on quasi-random versus pseudo-random samples from copulas are investigated in dimensions as large as 100 for the first time. Second, replicated validation MMDs, as well as Monte Carlo and quasi-Monte Carlo applications demonstrate the improved training of AGMMNs for a copula model implied by the 50 constituents of the S&P 500 index after deGARCHing. Last, both the latter dataset and 50 constituents of the FTSE 100 are used to demonstrate that the improved training of AGMMNs indeed translates to an improved model prediction.

The goal of the paper is to present two simple proofs of the Lagrange Inversion Formula for formal power series. Both proofs are non-external in the sense that they use concepts that do not go beyond the scope of basic tools of formal power series analysis.

High-resolution large-eddy simulations of decaying stratified and unstratified homogeneous turbulence are used to understand the mixing of passive scalars in stably stratified flows. Two passive scalar mixing layers, one in the vertical direction and the other in the transverse direction, are a model for a plume that is very large relative to the length scale of the velocity. In the transverse direction, the evolution of the passive scalar is broadly similar in the stratified and unstratified cases, although it does spread slightly faster when stratified. Also, the intensity of the scalar fluctuations is higher in the stratified case, and the turbulent/non-turbulent interface is more intermittent. In the vertical direction, though, the stratified case has almost no mixing because the stratification prevents large-scale stirring. Initially, the stratified passive layer grows until its width is proportional to the vertical integral length of the horizontal velocity, which is itself constrained to maintain the vertical Froude number order one. After this early growth, there is little additional spreading of the passive scalar. Modelling of the stratified scalar flux in the transverse direction is done effectively with a one-constant model if the mean profile is known, and a two-constant model if the profile shape must be assumed. In the latter case, the model is good only if the scalar is in quasi-equilibrium with the velocity field such that the length scale of the scalar can be scaled from the kinetic energy. In this study, the Prandtl number of the active and passive scalars is 0.7. It is anticipated that the reverse buoyancy flux resulting from higher Prandtl numbers will affect the passive scalar mixing.

Photothermal imaging is a powerful noncontact and nondestructive technique for subsurface inspection of composite materials, yet its performance is fundamentally limited by the diffusive and irreversible nature of heat transport, leading to severe image blurring and ambiguous depth interpretation. The concept of virtual waves provides a route to overcome this limitation by linking diffusion fields to propagating wave fields, but existing approaches are largely restricted to idealized impulsive excitation. Here, we propose a generalized virtual-wave photothermal tomography framework that extends the diffusion-to-wave transformation to arbitrary boundary excitations, including pulsed, harmonic, and chirped waveforms. Starting from the heat equation with a general source term, we derive a Fredholm integral mapping between the measured diffusion field and a virtual wave field governed by a wave equation, explicitly enforcing causality and thermodynamic irreversibility. The resulting ill-posed inverse problem is solved using ADMM or truncated SVD, depending on the excitation characteristics. Numerical and experimental results demonstrate that the proposed method converts blurred thermal responses into wave-like fields with clear wavefronts and reflections, enabling improved depth localization and tomographic reconstruction. Experiments on carbon fiber reinforced polymer samples with embedded defects show enhanced contrast, sharper boundaries, and more reliable depth interpretation compared with conventional thermographic techniques. This work establishes a unified and physically grounded framework for wave-based photothermal tomography under realistic excitation conditions.

This manuscript maps secure-supply and criticality risks for quantum technologies deployed in extreme environments, linking upstream critical minerals and materials (CMMs) to downstream system performance, continuity of security, and mission assurance. It adopts a reproducible "Critical Level I" screening method to identify materials whose supply concentration, essentiality, and limited mitigatability can create bottlenecks for quantum deployment. The analysis is structured around two use cases: (i) niobium as a key input for superconducting quantum computing and related manufacturing and toolchain dependencies; and (ii) space-qualified superconducting nanowire single-photon detectors (SNSPDs), alongside adjacent single-photon detector platforms such as SPADs, where radiation, thermal cycling, vibration, and electromagnetic interference can degrade device metrics and, in communications settings, threaten continuity of security. The manuscript further situates these dependencies within U.S.-China strategic competition over critical materials, refining capacity, export controls, and overseas mineral acquisitions, while also connecting them to standards-first governance, post-quantum cryptography migration, and the emerging security logic of quantum networking. It argues that static national critical-minerals lists are insufficient for mission-relevant quantum technology and proposes a dedicated Quantum Criticality and Critical Minerals (QCCM) dashboard as a living decision-support tool for tracking concentration, substitutability, qualification bottlenecks, stockpiling gaps, and geopolitical stress signals across quantum platforms. The paper concludes with implications for substitution, diversification, stockpiling, shielding, qualification-by-design, and standards-aligned governance to support secure, sustained, and mission-relevant quantum deployment.

We construct the generalized Komar charge of generic, non-linear theories of electrodynamics (NLED) in 4 dimensions coupled to Einstein gravity. The contribution of the dimensionful coupling constant present in all these theories is obtained by promoting it to a dynamical field which is forced to be constant on-shell by a Lagrange multiplier. We use this charge to derive a Smarr formula for asymptotically-flat black-hole and soliton solutions of these theories that includes the contribution of the coupling constant. Previously, this contribution had been found using homogeneity arguments. We test our results on a broad class of Einstein--NLED theories and analyze in detail the thermodynamics of the regular Bardeen black hole using the conservation of the generalized Komar charge to understand the regularity of regular black holes inside the event horizon.

The discovery of high-$T_c$ superconductivity in Ruddlesden-Popper nickelates has sparked substantial effort towards understanding unconventional electronic states beyond a traditional cuprate-like $d^9$ configurational ground state. An understanding of the interplay between magnetic ground states and multi-orbital physics is key for establishing a microscopic mechanism for superconductivity. In the bilayer nickelates, spin density wave (SDW) order is a prominent feature in the non-superconducting regime, yet its relation to superconducting pairing remains an open question. Here, we use resonant x-ray scattering to examine the existence of SDW order in superconducting bilayer nickelate thin films La$_2$PrNi$_2$O$_7$ (LPNO). Comparing superconducting and oxygen-deficient LPNO thin films, we find that superconductivity occurs in SDW-free, oxygen-stoichiometric regions, whereas oxygen-deficiency promotes SDW order, indicating phase segregation of SDW and superconductivity. Furthermore, Ni-$L_3$ and O-$K$ edge spectroscopy reveals distinct electronic structures - particularly along the $c$-axis - between the two regions. Our results identify oxygen stoichiometry as a key parameter controlling interlayer coupling and thus the electronic structure of bilayer nickelates. In concert with theory, we propose that a ligand hole primarily resides at the inter-bilayer apical oxygen, forming a robust interlayer five-spin polaron state, which serves as the ground state for superconducting bilayer nickelates.

Metallicity can play a significant role in massive binary evolution through its impact on the opacity within stellar interiors and wind-driven mass loss. In this work, we investigate how the double neutron star (DNS) delay time distribution (DTD) is shaped by the metallicity-dependent evolution of the helium star$-$NS progenitor system. Drawing from insights rooted in single and binary star physics, we argue that at a given metallicity, the stellar radius during the helium main-sequence sets a lower limit on the size of the DNS orbit at birth. We then perform population synthesis with the detailed binary evolution code POSYDON to illustrate the resulting DTD across a range of metallicities. Our results indicate that, independent of the common envelope efficiency and reasonable natal kicks, the majority of DNS mergers across metallicities occur typically no earlier than $\simeq 40\,\rm{Myr}$ after star formation and peaks strongly between $80-250\,\rm{Myr}$. Roughly $15\%$ of DNSs merge within 80 Myr, which may explain $r$-process enrichment in environments with brief star formation histories, while $\gtrsim 20\%$ merge on delay times $>1$Gyr, providing an explanation for short gamma-ray bursts in old, metal-poor galaxies. The shape of the DTD can be complex, with a metallicity-dependent split in the dominant formation channel imprinting a characteristic double-peaked structure. Although ideally oriented natal kicks can produce very short merging DNS, we find that the required kick magnitudes are inconsistent with observations. Our work has implications for assessing the contribution of DNS mergers to $r$-process enrichment and gamma-ray bursts/kilonovae transients across cosmic time.

We classify when a spin and a non-spin irreducible character of the alternating group have proportional 2-modular reductions. Equivalently, we classify when such a pair of characters are proportional on elements of odd order.

Fair cooperative multi-agent reinforcement learning (MARL) teams that maximize an egalitarian welfare are exploitable: a single self-interested agent free-rides on the surplus that fair agents forgo to raise the worst-off, and the known remedy is a centralized need-based allocator. We show that a decentralized defense becomes possible once contention is graded: when a contested resource still delivers a fraction $1-c$, a worst-off cooperator that contests a free-rider strictly improves on yielding, so leverage exists for every $c < 1$. We introduce CAN, a permutation-equivariant cross-attention policy over agents' observed behaviour that infers how many free-riders are present and responds proportionally -- turn-taking when none, contesting just enough when some. Trained against an adversarial league, CAN keeps best-response exploitability near the centralized oracle ($ρ\approx 1.2\text{--}1.5$ vs. $ρ= N$ unprotected) at essentially no efficiency cost, whereas the fair-MARL learners (GGF, FEN, SOTO) each collapse to an exploitable or wasteful extreme. Giving those objectives CAN's identical adversarial training does not rescue them, so the objective -- not adversarial training alone -- is what makes hardening possible. Against a committed (non-adaptive) defector, every learned defense including ours provides deterrence rather than immunity, weakening as the leverage $(1-c)/2$ vanishes. Across further environments and team sizes the same principle sets the scope: robustness holds exactly as far as the game's contest leverage reaches, and we map that boundary rather than claim to remove it.

Real-time dynamics of interacting electrons lies at the interface between quantum mechanics and non-equilibrium physics, governing the microscopic origin of ultrafast phenomena of molecules and nano-materials. Though neural network variational Monte Carlo has achieved unprecedented accuracy for stationary state calculations, its extension to real-time evolution remains challenging. In this work, we introduce the neural basis time-dependent variational Monte Carlo framework, which achieves stable and highly accurate simulations of electron dynamics. By constraining the time evolution to a compact, customized manifold spanned by the neural basis, we effectively bypass instability issues and achieve long-term stable evolution. Moreover, we demonstrate that this framework yields benchmark-quality accuracy in simulating the laser-driven dipole responses of the hydrogen atom and a stretched hydrogen molecule, and accurately extracts the dynamic polarizabilities of helium and beryllium atoms. Our work reveals the vast potential of neural network wavefunctions for accurately describing real-time electron dynamics and establishes a promising new route for first-principles simulations of complex, time-dependent electronic phenomena.

We present a compressed Gaussian likelihood for the Planck CMB low-$\ell$ E-mode polarization data, constructed from the SRoll2 likelihood which provides the tightest constraint on the reionization optical depth $τ$ to date. The non-Gaussian form of CMB low-$\ell$ TT and EE likelihoods makes them incompatible with Fisher matrix analyses that require an analytic Gaussian $χ^2$, such as the Fisher-bias formalism and Fisher forecasts. We show that the $χ^2$ of an offset log-normal likelihood takes a Gaussian form in the log-transformed power spectrum amplitudes, and can therefore serve as a proxy for the true Gaussian likelihood of this variable in Fisher matrix analyses, without any explicit change of variables. Building on this, we compress the SRoll2 likelihood into a small number of piecewise offset log-normal functions and validate it against the full SRoll2 likelihood via MCMC combined with Planck and ACT DR6 data, finding excellent agreement across all $Λ$CDM parameters and in extended cosmological models. We further demonstrate that Fisher matrix uncertainty estimates from our compressed likelihood agree well with the full MCMC posteriors. We release our compressed likelihood planck-gaussian-lowl, a lightweight Python package incorporating the compressed low-$\ell$ TT likelihood from previous work, allowing a straightforward incorporation of the Planck CMB low-$\ell$ data into any Gaussian-likelihood-based analysis. The package is publicly available at \href{https://github.com/nanoomlee/planck-gaussian-lowl}{github.com/nanoomlee/planck-gaussian-lowl}.

We present experimental findings from a study (N=99) examining how intellectual humility (IH), i.e., the metacognitive awareness of epistemic limitations, affects the evaluation of AI-generated health dialogues varying in scientific rigor. Participants were randomly assigned to evaluate one of three dialogues about exercise and mental health: scientifically accurate, moderately pseudoscientific, or strongly pseudoscientific. Results reveal that IH functions as a selective cognitive filter. Individuals with higher humility scores rated pseudoscientific content as significantly less credible, while showing no correlation with credibility assessments of accurate content. Crucially, humility did not predict the ability to identify AI as the source of dialogues, suggesting that epistemic vigilance operates on content quality rather than source attribution. We interpret these findings through an evolutionary lens, proposing that IH represents an ancestral adaptation for navigating informationally uncertain environments. It remains effective at detecting exploitation attempts in AI-generated content, despite humans lacking evolved mechanisms for detecting AI sources. The study contributes to understanding how foundation models might improve or undermine human epistemic defenses, especially in health communication contexts.

Ollivier-Ricci curvature of an edge (x,y) is defined by comparing the distance taken to transport from neighbors of x to neighbors of y. It is a structural measure that has been studied in many fields such as community detection and deep neural networks. However, high computational complexity or error limits its application in large scale-free graphs. This paper proposes an optimal transport principle to minimize the distance by 3,4,5-cycles that include the edge (x,y), and designs a curvature calculation approach named Curvature in Cycle Overlap Mode (CCOM). In this approach, a greedy and pruning algorithm is proposed to approximate the optimal transport principle. We theoretically and experimentally verified that our approach CCOM can significantly improve the accuracy of the curvature on real-world networks with low time consumption. In addition, we compared CCOM with baseline approximation approaches in community detection tasks using the same curvature-based framework, and experimentally confirmed the effectiveness of CCOM on large scale-free graphs.

Fuzzy deduplication is key to constructing large language model training corpora. However, classic Locality-Sensitive Hashing (LSH) pipelines scale poorly as corpora grow and are ill-suited to continuous ingestion. The main issue is that each new document batch must be checked against the admitted corpus before insertion. As the corpus grows, the LSH buckets grow: each query can hit several large buckets and must scan the returned candidates. To solve this problem, we present RAD (Retrieval-Augmented Deduplication), an online fuzzy deduplication system that delivers both high recall and throughput for evolving datasets. RAD maintains an incrementally updated HNSW index over admitted documents, retrieving a small, high-quality candidate neighborhood for each incoming document instead of repeatedly re-scanning the accumulated corpus. RAD is the first online fuzzy deduplication system to use HNSW, leading to stable throughput as datasets grow. However, it is not easy to maintain high recall when using HNSW-style indexes. The core issue is the distance metric between graph nodes. Jaccard similarity, the metric used for fuzzy deduplication, yields low recall when applied out-of-the-box with an HNSW index. It leads to distance score crowding, making graph traversal unreliable within a bounded number of steps. RAD addresses this with a bitmap representation that provides a more discriminative, Jaccard-aligned signal during HNSW search. Across four LLM-scale datasets (LM1B, C4, RealNews, and Common Crawl), RAD preserves the scaling trajectory needed for online fuzzy deduplication: at 30M documents, it maintains 0.94-0.97 recall relative to state-of-the-art LSH solutions, and delivers up to an 8x throughput increase.

A unified, closed-form analytical PI/PID tuning method is presented for all-pole plants up to third order that yields a strictly monotonic (zero-overshoot) step response with minimum settling time. The design target is the binomial closed loop p^n/(s+p)^n, which is monotonic with robustness depending only on the order n. Because a fixed PI/PID cannot assign the closed-loop poles and the controller zeros independently, realizing this target exactly requires the controller zeros to be cancelled, which forces the controller numerator to divide the plant denominator. It follows that an exact, real-gained solution exists for any stable plant precisely up to second order with a PI controller and third order with a PID controller; beyond that the residual binomial factor acquires a complex pair of damping sqrt(3)/2, which a generic plant does not contain. Explicit gains are derived for first-order plants (PI), second-order plants with real and complex poles (PI and PID), and third-order plants with three real poles or one real pole plus a complex pair (PID). The freedom of the coincident designs is shown to be bounded: a quadratic nonnegativity condition gives the exact window of the design pole for strict monotonicity, which collapses at the pole-ratio-2 changeover for real poles and is nonempty for damping ratios above approximately 0.443 for complex poles. Monotonicity guarantees Mt = 1, hence Ms <= 2, phase margin >= 60 degrees, and gain margin >= 6 dB, tightening to universal constants for the binomial family. Load-disturbance attenuation obeys IAEd = 1/Ki, making the cost of cancellation explicit, and comparisons with SIMC, the CHR zero-overshoot rule, and deadbeat-fitted explicit formulas quantify the trade: at matched maximum sensitivity the proposed design settles faster than SIMC on the third-order example, with markedly lower controller gains and peak control effort.

This article is devoted to the study of the evolution of the properties of nucleons bound in nuclear matter within the framework of the chiral confining model. The in-medium nucleon trial states (either localized factorized wave functions or momentum-projected states) are determined by imposing the von Laue stability condition, according to the formal results established in a preliminary companion paper (labeled as I). The main results concern the response of the composite nucleon to the scalar field, as well as the respective roles of confinement and chiral symmetry breaking in the evolution of the in-medium nucleon mass. This evolution governs the repulsive three-body forces required for the nuclear saturation mechanism. We also analyze the modification of the energy density distribution and the pressure distribution inside the in-medium nucleon. We also draw some perspectives concerning the mapping between bound nucleon properties and the equation of state of dense matter as realized in the deep interior of neutron stars.

We discuss the issue of the mechanical stability of the nucleon within a class of models in which massive constituent quarks are subject to a confining potential and are coupled to a surrounding pion cloud enveloping the quark core. The nucleon trial states (either localized factorized wave functions or momentum-projected states) are determined by imposing the von Laue stability condition. This article is primarily devoted to the formal aspects related to the detailed expressions for the total energy (mass), the average pressure, the energy density and the pressure distribution inside the nucleon. It will be accompanied by a complementary article addressing the evolution of nucleon properties with density, associated with the restoration of chiral symmetry.

We investigate Kerr-Schild-AdS geometries in quadratic f(R) gravity without imposing the constant-curvature condition R=R_0 a priori. We show that the field equations dynamically enforce constant scalar curvature and uniquely select the Kerr--AdS family of solutions. Thus, quadratic f(R) gravity admits no Kerr--AdS solutions beyond the Einstein branch, establishing a no-go theorem for this class of geometries.

Identifying crack generations from microscopic images of hierarchical crack patterns is challenging due to the lack of temporal information and sensitivity to image boundaries. Existing algorithms often fragment individual cracks or lose stability when the observed fragment is shifted. We propose a method that reduces the classification problem to topological sorting of a directed acyclic graph (descendant$\to$parent), built from T-junctions and nearly collinear edges. Sequential removal of leaf vertices assigns generation numbers starting from the youngest. On 100 computer-generated networks, our method correctly classifies $\approx 70$\% of cracks at a window size of only three mean edge lengths, whereas a conventional approach that starts from primary cracks drops nearly to zero. The classification is highly stable against reasonable shifts of image boundaries but remains limited to strictly hierarchical networks.