Papers

Collision cross section (CCS), derived from ion mobility mass spectrometry, is a common descriptor for molecular annotation. Prediction is challenging for machine learning models because it reflects the size, shape, and ionization state of a gas-phase molecular ion. Most predictors either ignore explicit 3D structure or treat adduct identity as a late categorical feature, which limits their ability to capture adduct-dependent geometric effects. We present GRACE (Geometric Residual Adduct Conditioning via Early-fusion), a 3D CCS predictor that adapts a pretrained molecular geometry encoder using geometric residual adduct conditioning via early fusion. GRACE combines two inductive biases: a residual objective relative to an adduct-aware physical descriptor baseline and adduct conditioning within the encoder via a learned adduct token and low-rank attention adapters. We evaluate the model on a curated set of over 9,000 experimental molecule-adduct CCS records with random, scaffold, and adduct-sensitive splits designed to separate interpolation, scaffold generalization, and adduct-driven generalization. GRACE achieves the best mean percentage difference among the evaluated learned models on all three splits: 1.67% on the random split, 2.11% on the scaffold split, and 2.36% on the adduct-sensitive split. Diagnostic analyses suggest that residual learning stabilizes training by removing the dominant mass-CCS trend, while early fusion improves adduct-sensitive prediction relative to late fusion. Across four independent external test sets, GRACE shows consistently lower error than the other evaluated models. On a held-out set, GRACE also attains the lowest mean percent difference when compared with four previously reported physics-based workflows. These results support residual learning and encoder-level adduct conditioning as practical inductive biases for fast, accurate CCS prediction.

The growing use of electroencephalography (EEG) motivates automated analysis that is accurate, transferable, and deployable on constrained hardware. Recent EEG foundation models learn general representations from large-scale pretraining, but their size and computational cost limit edge and wearable deployment. We introduce BRIDGE-EEG, an efficient multi-task EEG classification pipeline that preserves the benefits of pretraining while reducing model size. A unified preprocessing scheme maps heterogeneous recordings with different channel counts, montages, and sampling rates to a device-agnostic 62-channel time–frequency representation. We pretrain an SE-ResNet18 teacher (11.84 M parameters) with SimCLR on unlabeled EEG from five heterogeneous datasets, then compress it into SE-ResNet8 (1.56 M) and SE-ResNet4 (0.48 M) students using task-agnostic and task-specific distillation. We evaluate six benchmarks spanning abnormality detection, motor imagery, and emotion recognition. For abnormality detection and emotion recognition, the students achieve accuracy comparable to or better than several recent EEG foundation models with 10–1,000$\times$ more parameters. Motor imagery shows a remaining representation gap, highlighting the importance of pretraining diversity. Inference profiling on a server GPU, desktop CPU, and NVIDIA Jetson Orin Nano shows up to 3.0$\times$ lower edge energy per inference (15.64 mJ vs. 46.67 mJ). The compact models further support future deployment on MCU-class wearables.

Characterizing two-dimensional (2D) quantum materials by optical microscopy requires localizing exfoliated flakes and determining their layer thickness from subtle optical contrast and interference color to select suitable flakes for device fabrication. However, models face synthetic-to-real domain shifts and variation across materials, substrates, laboratories, and imaging conditions. We present QuPAINT, a physics-aware multimodal framework for transferable quantum flake characterization. The Synthetic Materials Framework (Synthia) generates diverse synthetic microscopy images while preserving layer-dependent optical behavior. Using these images, we construct QMat-Instruct, a multimodal instruction dataset with image-specific reasoning traces generated from verified annotations and constrained to observable optical cues. QuPAINT integrates these signals through Physics-Informed Attention (PIA), which injects substrate-relative optical priors into the visual representation to support grounded multimodal reasoning. For evaluation, we introduce QF-Bench, to our knowledge, the largest real-world benchmark for this problem, spanning diverse microscopy and substrate conditions. Using its verified annotations, we study counting, visual grounding, reasoning quality, confidence calibration, and transfer to an unseen material. QuPAINT-8B substantially outperforms prior methods and establishes state-of-the-art performance for both general and monolayer flake detection. Additional experiments show that image-grounded supervision improves strict spatial grounding and confidence calibration while preserving robust general flake detection on the unseen material.

Comparing and selecting task-oriented LLM agents increasingly relies on a low-cost offline evaluation gate: persona-driven LLM user-simulators converse with each candidate, an LLM-as-a-judge scores the transcripts, and the higher-scoring agent is promoted. We introduce GAUGE, a reusable offline protocol that measures whether this gate's ranking matches a grounded verifiable reward across 25 agents from six providers on the $\tau^2$-bench and SimulatorArena benchmarks, separating two kinds of evaluation validity that release practices conflate: ranking validity and construct validity. First, a satisfaction-success gap: satisfaction carries essentially no information about task success, as conversations rated satisfied by our blind panel are decorrelated from actual success, with 57.5% of them failing the customer's task, a pattern consistent across five rater populations, both benchmarks, and every subjective dimension we rated. Second, while the gate's ranking is robust across the broad capability span, it loses resolution among the near-equal strong agents: this decision-disagreement rate jumps from $<$1% on wide-reward pairs to 31% on close pairs. The gate is thus human-validated yet mis-anchored. As a remedy, we propose a calibrate-then-trust cadence in which a judge-free completion bit is a zero-cost tripwire for truncation regressions.

Experimental TCAD calibration is essential for predictive technology modeling of emerging oxide semiconductor transistors. However, it remains time-consuming and expert dependent because of model ambiguity. Multiple physical models and parameter sets can reproduce the same measured transfer characteristics, while local fitting alone cannot uniquely identify the underlying device physics. We present the first demonstration of an agentic TCAD calibration workflow for a fabricated bottom-gate In–W–O (BG-IWO) transistor. Starting from the measured transfer curve and device information, the workflow uses measurement–TCAD residuals and local sensitivity tests to select bounded parameter corrections or evaluate additional physical models, and accept only updates that improve device metrics. The LLM agent orchestrates the workflow, while Sentaurus governs the device physics. For the 2%-W reference device, five agent-suggested updates yield a fixed calibrated model, reducing the multi-metric device objective $J$ by 14.3$\times$. Maximum $V_{\mathrm{th}}$/$I_{\mathrm{on}}$ errors are 36.1~mV/0.022 decade for varying-drain-bias tests and 46.2~mV/0.062 decade for varying-channel-length tests, demonstrating model transferability across bias and geometry rather than a local parameter fit. W-composition tests provide process-sensitive insight. This agentic workflow provides a faster route to model development for emerging device technologies.

Algorithmic recourse methods typically assume that a predictive model has access to all features of an individual. In practice, decisions are often made with partial information, because features are costly to acquire. Active feature acquisition addresses cost-constrained prediction, but existing methods are explanation-agnostic: prior work provides explanations only after acquiring additional features, rather than using explanations to drive acquisition. This work flips that and treats algorithmic recourse and feature acquisition jointly. We use Markov Blanket theory to unify counterfactual, semifactual, and alterfactual explanations and to characterize how available recourse grows as features are acquired. Building on this framework, we propose an Explanation-Driven Feature Acquisition (EDFA) method that selects features by explanatory value per unit cost. The framework is further extended with distribution-free validity guarantees for recourse issued from partial information, which signal trustworthy, lower-cost recourse, along with a lower bound on the calibration data required to certify them. Experiments on 7 publicly available datasets with neural network-based predictive models show that EDFA acquires substantially fewer features than state-of-the-art AFA baselines while maintaining comparable accuracy and yielding more decision-relevant, actionable recourse. The implementation is available on GitHub.

Transportation agencies need pedestrian volume estimates across entire road networks to prioritize safety investments, yet manual counts are expensive and cover only a small share of intersections. We present a machine learning pipeline that predicts 2-hour PM peak pedestrian volume at 101 urban intersections in Portland, Oregon, from built-environment, land-use, and street-network features drawn from open GIS data. Starting from the Negative Binomial GLM used in practice, we add feature selection, count-aware gradient boosting, and repeated cross-validation, selecting one configuration by a combined rank over RMSE, MAPE, and SMAPE across four cross-validation strategies. The winner, a histogram-based gradient boosting model with Poisson loss and L1 Lasso feature selection, reduces cross-validated RMSE by 12% over the GLM baseline (89.8 to 78.7) and holdout RMSE by 19% (108.0 to 87.9). Code is released on GitHub.

Existing evaluations of concept unlearning in text-to-image (T2I) diffusion models primarily rely on attack success rates obtained through automated adversarial prompt search. However, these metrics provide only empirical evidence over a finite set of queries and leave residual leakage over the broader prompt space largely unquantified. This limitation can lead to overestimating unlearning effectiveness and underestimating safety risks. To address this gap, we introduce a novel certification framework for T2I concept unlearning that provides high-confidence guarantees with bounded error on residual concept leakage. Our approach combines statistical certification with worst-case analysis along concept-relevant embedding directions to derive explicit upper bounds on leakage probability under user-specified confidence levels. We evaluate our framework across three major concept categories namely NSFW content, artistic styles, and celebrity identities, and six state-of-the-art unlearning methods. Certified leakage bounds consistently exceed standard attack success rates by 16.2%, uncovering substantial residual risks missed by existing evaluation protocols. Crucially, our results demonstrate that empirical attack-based evaluations can significantly underestimate residual leakage and establish certification as a necessary complement for reliable auditing of concept unlearning in T2I diffusion models.

We present a framework for tracking topological features directly within continuous implicit models. Such models, including implicit neural representations (INRs) and multivariate functional approximations (MFAs), are increasingly adopted to represent scientific data without the resolution constraints of discrete grids. They offer compact, smooth, and differentiable representations of complex fields, enabling new opportunities for high-performance data storage, reconstruction, and analysis. Given a continuous implicit model, our method tracks the evolution of critical points by querying the model and its derivatives, thereby eliminating the need to resample onto a grid. This approach enables faithful feature tracking while avoiding discretization-induced artifacts such as aliasing. We demonstrate the generality of our framework across a range of implicit representations, including analytic functions, MFAs, and INRs, and show that it produces smooth, coherent critical point trajectories. By enabling feature tracking directly on continuous representations, our method supports a new class of feature-driven visualization workflows centered on implicit models.

AI-native 6G networks have brought Intent-Based Networking (IBN) to the forefront, enabling high-level goals to be translated into network configurations. However, this abstraction opens new attack surfaces, primarily adversarial intent injection, where malicious policies are disguised within benign intent flows. The detection of attack instances might become significantly more difficult if the adversaries adopt a stealthy mode of malicious intent injection. With all these in mind, we first define a fine-grained threat model that facilitates the threat of malicious intent injection in an AI-native network. Alongside, we investigate four malicious intent injection strategies$-$ stealth-mode, random distribution, increasing frequency, and decreasing frequency- and propose a dual-path detection framework: (i) a CNN using TF-IDF features for supervised malicious intent detection, and (ii) an AutoEncoder trained exclusively on benign data for one-class malicious intent detection. Our evaluation demonstrates strong detection performance, with accuracy improving to 0.97 (~9% gain) and F1-score to 0.98 (~36% gain) over the state-of-the-art baseline.

Measuring the preferences of human users remains a fundamental challenge of AI alignment. Existing elicitation approaches struggle to efficiently discover multidimensional preferences or accurately ground these inferences in domain knowledge. To address this, we introduce GUIDE, an LLM-driven elicitation architecture that infers user preferences through conversations by combining Bayesian adaptive sampling for question selection and symbolic representation learning to initialize domain-specific preference models. GUIDE generalizes adaptive sampling to diverse elicitation questions through an extensible type system of transforms on a parameterized preference state. GUIDE produces domain-specific preference representations through an initialization process using symbolic rule-based learning to capture world knowledge and set priors over preference dimensions grounded in data about decision alternatives. The architecture provides observability and steerability to facilitate deployment and analyze elicitation processes. In silico experiments on investment portfolio optimization demonstrate that GUIDE improves cold-start and minimizes recommendation regret consistently within early elicitation interactions across user personas compared to prior work, LLM-only baselines, and ablated GUIDE versions.

Low-Rank Adaptation (LoRA) is an effective approach for adapting large pretrained models by learning low-rank weight updates. In practice, the LoRA rank is used to control an adapter's parameter budget and representational capacity. We show that this view is incomplete: while the nominal rank determines the representational capacity, the optimizer shapes how much of that capacity is used in the induced weight-space updates. In a case study of GPT-2 adaptation with LoRA, we observe a strong rank-dependent optimizer effect. Despite using the same nominal rank, AdamW often produces per-step updates with concentrated singular spectra and low effective rank, whereas Muon uses a richer set of directions and benefits more consistently from increasing LoRA rank. These observations motivate ISO-LoRA, an optimizer that couples the LoRA factor updates through spectral descent on the induced tangent perturbation in weight space. ISO-LoRA promotes updates that distribute energy more evenly across singular directions, improving rank utilization while preserving compatibility with the LoRA parameterization. We complement this design with theoretical guarantees showing that ISO-LoRA can achieve higher effective rank than standard factor-wise optimizers through a one-step analysis under a stylized spiked-gradient model. We validate this design on language-model adaptation across 0.1B-7B-parameter models, where ISO-LoRA improves effective rank and downstream performance, with the strongest gains at moderate-to-large LoRA ranks. Our results highlight rank utilization as a key factor in LoRA optimization and suggest that optimizer design offers an important path toward stronger parameter-efficient adaptation.

Stochastic gradient descent (SGD) with gradient clipping and additive noise has become a standard technique for training machine learning models, particularly in applications requiring robustness or privacy guarantees. However, clipping introduces a bias in stochastic gradients, while additive noise introduces additional variance, making the long-run behaviour of individual optimization trajectories difficult to characterize. In this work, we prove that SGD with clipping and additive Gaussian noise (SGD-CN) converges almost surely (a.s.) under smoothness and uniformly bounded stochastic-gradient noise assumptions, provided the step sizes satisfy some standard decaying conditions. Our analysis extends to momentum variants such as the stochastic heavy ball and Nesterov's accelerated gradient, where we show that careful energy constructions yield similar guarantees. These results provide stronger theoretical foundations for understanding the pathwise behaviour of clipped stochastic gradient methods and suggest that, despite the bias and noise introduced by clipping and perturbation, the algorithm remains stable in both convex and nonconvex regimes.

Outliers are important for stress-testing algorithms and understanding system behaviour under rare conditions. Despite being commonly described as low-likelihood events, existing generative approaches rarely control likelihood explicitly. In this work, we introduce a measure-theoretic notion of outliers based on the distribution of log-likelihood values, which is guaranteed to assign higher probability mass to low-likelihood events with a specifiable magnitude. Building on this formulation, we derive how likelihood reweighting modifies the diffusion score and use this relation to motivate a controlled modification of the reverse-time dynamics. In particular, likelihood reweighting implies a scaling of the score function with a control term derived from the Radon-Nikodym derivative of the likelihood distributions. Correspondingly, the updated score function can be obtained with no retraining of the diffusion model. We exploit the Ornstein-Uhlenbeck semigroup underlying diffusion models to motivate an exponentially interpolated controller which approximates the true control. Experiments demonstrate controlled generation of low-likelihood samples while remaining consistent with the data geometry.

The prevailing assumption in applied machine learning is that progress on consequential quantitative decisions such as pricing risk, allocating capital, triaging patients, or containing a network intrusion will follow from progress in large language models (LLMs). A language model is trained on a representation of the world that was produced by human description; description is a lossy encoding of the quantitative record, and the loss is irreversible: no downstream model, at any scale, can recover from a description what the description did not encode. We formalize this as a property of the representation on which a model is trained rather than of the model capacity, and we identify three further properties that consequential settings demand of a model and that a language substrate cannot supply by construction: reproducibility, lineage from every output back to the source records that produced. it, and calibrated uncertainty. We argue that these properties define a distinct model class, which we call the Large Quantitative Model (LQM).

Detecting whether a transmitter is actively communicating with a receiver is a fundamental problem in molecular communications. A spherical receiver may measure not only the number and arrival times of absorbed molecules, but also their absorption locations on the receiver surface. These locations contain a directional signature that is lost in count-only detection. In this letter, we develop a molecular signal detector based on a Legendre polynomial expansion of the receiver-surface hit density and extend it to a Legendre-based Viterbi sequence detector. By exploiting the surface-level directional signature of molecular arrivals, the proposed detectors provide geometry-aware, efficient, and better-performing alternatives to count-only detection.

Hierarchical correlations are a universal feature of any realistic model of data, and the question of how associative memory models may learn these correlations and generalize beyond them to construct new sensible images is an important step towards understanding more complex modern architectures such as diffusion models. We consider a hierarchical model for memories which are sampled and stored in a dense Hopfield network with polynomial activation. We analytically derive conditions for each level of this hierarchy to be locally stable - that is they are local energy minima. We use prototype reconstruction as a minimal model of generalization and we find that it takes only a quasi-polynomial amount of information to generalize beyond particular memories and even particular groups in the hierarchy. We observe a qualitatively analogous phase diagram in the number of memories, sharpness of the activation function (polynomial degree) for data from Fashion-MNIST.

Vision-Language-Action (VLA) models show high robotic manipulation capabilities via a two-stage design: a Vision-Language Model (VLM) stage followed by an Action Diffusion Transformer (ADiT) stage. Since robots must meet strict Service-Level Objectives (SLOs) for safety, VLA inference is inherently latency-critical. Meeting these SLOs requires high-end GPUs, yet weight, cost, and power constraints preclude integrating such GPUs on-robot. Prior works offload VLA inference to edge servers that serve many robots on VLA models. However, current VLA systems lack support for multi-request, multi-model execution on a multi-GPU server under SLOs, while existing serving systems for multi-stage models are optimized for throughput and stage disaggregation across separate GPUs, which are ill-suited for the millisecond-scale stages of VLA models. We design Robion, the first VLA serving and management system for multi-robot, multi-model requests on multi-GPU edge servers that meets SLOs. Our serving engine disaggregates the VLM and ADiT stages within a GPU via two streams, dynamically restricting the SMs on VLM stream so ADiT always finds SMs to run alongside it, and co-locates multiple models by sharing these streams across them, prioritizing requests by least remaining SLO time. Our management engine enables flexible model placements on multi-GPU servers, and integrates an intelligent traffic controller that maximizes per-model batching under the chosen placement while bounding each GPU's load to meet SLOs. For individual models, Robion serves on average 6.7$\times$ and 1.5$\times$ higher robot load within 98% SLO attainment over vLLM-Omni, the most widely used multi-stage serving system, and Monolithic, which runs VLM and ADiT as a single pipeline, respectively. In a large-scale experiment of serving 8 different models on a 4-GPU server, Robion can serve up to 64 robots within 98% SLO attainment.

Channel simulation has recently emerged as a useful component in machine learning systems where samples from a prescribed probability distribution are to be compressed. Yet, general channel simulation algorithms often suffer from high computational costs, random stopping times or, in the worst case, can require generating an infinite number of shared random samples. We introduce a scheme for both exact and approximate simulation of discrete-to-continuous channels which conversely uses a fixed number of random samples, and therefore has a runtime independent of the channel and the input. Unlike existing channel simulation schemes which generate a sequence of independent samples from a proposal distribution, our approach generates one sample, or alternatively a fixed number of samples, from each potential target distribution. We then apply a latent permutation to the samples before performing sample selection using an exponential race. Our scheme provides a flexible tradeoff between the number of generated samples and the compression rate. Using polar and multilevel coding, we scale our approach to handle long blocklengths in $O(n \log n)$ time in order to benefit from reduced per-symbol overhead. We conclude by demonstrating applications to variable-rate compression with stochastic VQ-VAEs and communication-efficient differentially private distributed mean estimation via exact simulation of the Gaussian mechanism.

Beam-induced background (BIB) from muon decay products will be an overwhelming and unavoidable background at a future Muon Collider. In order to develop robust event reconstruction algorithms, we need large amounts of accurate BIB simulation to test on. BIB simulation is currently compute-limited: the simulated sample presently used for BIB overlay, which statistically represents approximately $10\%$ of a single unique event's worth of simulated BIB, requires on the order of $10^6$ HS23$\cdot$hours to generate and occupies approximately $100$ GB on disk. In this work, we develop the first machine learning models for fast BIB generation in tracking detectors. We consider two classes of architectures: a slower but higher-fidelity tabular diffusion model, and a faster but lower fidelity circular spline flow model. We find that both classes of architectures produce BIB hits and tracks that closely resemble those of available full simulation hits and tracks, and that the machine learning models can produce BIB in over an order of magnitude less time than what is needed to produce full simulation BIB. We release the model weights with the paper so that the Muon Collider community can use these fast BIB hits for future R&D.

Optimal transport gives structured data a geometry, but exact evaluation is costly in large pairwise analyses that exploit relationships among distances. Learned surrogates are faster, but need not preserve this metric structure. We develop a Deep Sets surrogate for OT between variable-size weighted point clouds that enforces non-negativity, exchange symmetry, and zero self-distance, leaving the triangle inequality unconstrained. Applied to the Energy Mover's Distance between collider events in a particle physics application, the Metric-Aware Particle Flow Network achieves percent-level mean absolute percentage error while significantly improving inference throughput over other exact and approximate methods surveyed. The architectural constraints are found to improve properties that are not explicitly enforced: across $10^6$ held-out event triplets, triangle-inequality violations fall from 199 for a matched unconstrained network to 2, and the maximum from 149.5 to 5.8 GeV. These results demonstrate that targeted inductive biases can yield fast neural surrogates with substantially improved geometric fidelity.

A key subtask of quantum error correction is to extract a syndrome that, if nontrivial, signals an error. The number of possible ways to extract a syndrome grows exponentially with the syndrome size, and these implementations vary greatly in fault tolerance, as measured by their logical error rates. This creates a natural search problem: find an implementation with a low logical error rate. Previous work solves this problem but sacrifices either solution quality or scalability. In this paper, we use reinforcement learning and importance sampling to outperform previous work at all scales. Compared with the state of the art automatic scheduling tools AlphaSyndrome and PropHunt, our tool reduces the logical error rate by 25.9% and 71.7% on average, respectively, culminating with a reduction of 97.8% for a surface code with distance 15.

Railway engineers need simulation models that predict vehicle responses across operating scenarios that cannot be tested exhaustively. Agreement with representative measurements provides essential evidence, but calibration at a limited set of conditions does not guarantee accuracy elsewhere. We present a multifidelity railway-bogie response-correction method that treats multibody simulation histories as low-fidelity information and roller-rig measurements as high-fidelity evidence. This method combines an experiment-anchored fidelity assignment with physics-informed discrepancy learning for multichannel bogie-response histories. A time-delay neural network (TDNN) represents the condition-dependent simulation trend, and development-fitted amplitude alignment defines the low-fidelity baseline. A residual-correction network then models the reproducible response component not explained by this baseline and adds it to the baseline. An effective dynamic-balance equation constrains the learned discrepancy by representing differences in inertia, damping, stiffness, and external forcing between the simulated and physical systems. The training objective combines this constraint with residual matching, temporal smoothness, and a combined channel-2 acceleration loss selected using displacement-acceleration consistency evidence. For the evaluated reconstruction case, the corrected response gives a mean coefficient of determination of 0.8197, a mean normalized root-mean-square error (NRMSE) of 4.6055 %, and a mean normalized mean absolute error (NMAE) of 1.9297 %. These results provide initial evidence of accurate response prediction at the held-out 385 km/h condition.

Self-triggered reinforcement learning control (RL-STC) learns the sparsest control schedule that preserves Lyapunov-decreasing stability under a Run-Time Assurance (RTA) override. We invert this: an adversarial RL agent learns the sparsest jamming or Denial-of-Service (DoS) schedule that destabilizes the closed loop, with a Lyapunov-increase admissibility predicate mirroring the defender's safety certificate. We prove a plant-property lower bound on the minimum jam count required for an immediate hold-last medium-access-control adversary to force a crash against a self-triggered controller (STC) satisfying a Lyapunov contract, and recover a certificate-level analog of the consecutive-grouping optimality of prior count-budget DoS scheduling as a corollary. This extends the DoS-scheduling count-budget analysis from periodic and linear-time-invariant to STC controllers. Empirically, we train against four fixed defenders per plant (one Linear Quadratic Regulator (LQR) and three RL-STC) on Pendulum, CartPole, and Quadrotor2D. The learned adversary is the only adversary that crashes every defender on every plant at $100\%$: greedy misses Quadrotor2D LQR on $42\%$ of episodes and periodic misses Pendulum LQR on $97\%$. On jam-time-per-failure it beats baselines by up to $2.8\times$, and shows its widest absolute margin on Quadrotor2D LQR. Robustness ablations show that Gaussian observation noise exceeding the initial-state magnitude and position-only observation both preserve $100\%$ failure rate and keep the learned adversary strictly ahead of both baselines on jam-time-per-failure.

Safe offline reinforcement learning assumes a cost function on every transition. We ask what remains possible when safety can be judged only by comparing short clips and occasionally asking whether an episode exceeded its budget. Certified safety curation answers with a filter-then-clone pipeline: a state-only value trained from segment comparisons scores whole trajectories, Learn-then-Test calibration certifies a selection threshold under a distribution-free $(\alpha, \delta)$ bound on the unsafe fraction of the selection, and behavior cloning follows. What is certified is the training set, not the policy, whose cost we report rather than bound. Where no threshold attains the target, the procedure refuses. We are not aware of prior work certifying the composition of a training set for offline RL or imitation. Oracle controls justify the design: reweighting individual transitions fails even with an exact value, so the value selects whole trajectories. The policies satisfy the cost budget on twelve of fifteen DSRL tasks, matching a clone of the ground-truth safe subset, which needs a label on every trajectory. Retrained on the certified selection, the strongest full-label method becomes safe where no setting of its own cost target rescues it. Refusal is predictable: the certificate's probability has a closed form in the purity of the pool's top quantile, which the calibration sample estimates and through which the scorer enters.

In small-scale binary sentiment classification scenarios, factors such as negation, contrastive shifts, and cross-word dependencies lead to the non-linear coupling of sentiment cues, making it difficult for conventional lightweight models to fully capture the contextual relationships between tokens. To address this issue, we propose a model named QTrans, which uses parameterized quantum circuits to construct query, key, and value features and derives attention coefficients from Gaussian distances between quantum measurements. By further integrating a quantum feed-forward neural network, residual connections, and layer normalization, the model establishes an end-to-end trainable quantum-classical hybrid framework for sentiment classification. Experimental results on the MR, CR, and MPQA datasets show that QTrans achieves test accuracies of 72.13%, 69.51%, and 63.45%, respectively, representing improvements of 2.88, 3.17, and 3.79 percentage points over the best-performing classical baselines for each dataset. Overall, QTrans expands the application of parameterized quantum circuits in lightweight sentiment analysis and lays an experimental foundation for further research into quantum multi-head self-attention for modeling textual relationships.

We propose a learning procedure for system identification in interacting particle systems from single-snapshot observations of collective behaviors, unlike existing approaches that rely on observations of trajectories. This setting leads to a fundamentally ill-posed inverse problem, which we solve by using a regularization strategy based on the empirical distribution of observed configurations, drawn from different, unobserved initial conditions. We test our learning procedure on a variety of representative models with steady-state and quasi-stationary patterns, where collective behaviors encode implicit information about the interaction mechanisms, demonstrating that our approach enables stable and accurate recovery of the underlying interaction laws, leading to faithful reproduction of the collective behavior, and in many cases even of the dynamics leading up to it.

LLMs are increasingly used as automated judges for model training and evaluation, yet individual judges exhibit systematic biases that undermine reliability. Much of prior work has studied biases in pairwise LLM-as-a-judge settings; in this paper, we focus on absolute scoring tasks, which mirror more realistic use cases. Across four benchmarks and six models (36 judge-examinee pairs), we show that a model's task accuracy strongly predicts its judging accuracy (Pearson $r \geq 0.90$ on most models) and inversely predicts its directional bias ($r \leq -0.83$), but that accuracy alone does not ensure fair evaluation: more capable examinee models consistently receive more lenient judgments from all judges ($r \geq 0.83$). To address this, we propose calibrated weighted majority voting (WMV), an ensemble evaluation method that aggregates multiple LLM judges weighted by online estimates of their false-positive and false-negative rates. We introduce a disagreement-based estimator that derives these error rates purely from inter-judge agreement patterns, requiring no ground-truth labels or task metadata. In a simulated experiment with shifting task distributions, our label-free WMV tracks an oracle with perfect error-rate knowledge to within 0.5 percentage points on average, outperforming both individual judges and unweighted majority voting. These results demonstrate that principled multi-judge calibration can simultaneously improve accuracy and correct for systematic leniency without requiring labeled data, offering a scalable path to reliable automated evaluation as model capabilities increase.

Generalization under distribution shift remains a core challenge in modern machine learning, yet existing learning bound theory is limited to narrow, idealized settings and is non-estimable from samples. In this paper, we bridge the gap between theory and practical applications. We first show that existing definition of concept shift breaks when the source and target supports mismatch. Leveraging entropic optimal transport, we propose a key notion: $\gamma^{*}\!$-concept shifts, and derive a general error bound unifying covariate and $\gamma^{*}\!$-concept shifts, which applies to broad loss functions, label spaces, and stochastic labeling. We further develop estimators for these shifts with concentration guarantees, and the DataShifts algorithm, which can quantify distribution shifts and estimate the error bound in most applications - a rigorous and general tool for analyzing learning error under distribution shift.

As the supply of human-written text is exhausted, it has become standard practice to repeat language model training data. Prior work has studied data repetition for densely activated Transformers, but the effects of data repetition remains largely unexplored for recently dominant sparse architectures such as Mixture-of-Experts (MoE), despite their increased compute efficiency. We vary data repetition rates across single- and multi-domain data mixes, and across MoE settings, including expert count and granularity. We consistently find, for models ranging from 80M to 1B active (8.5B total) parameters, that MoEs degrade more rapidly under data repetition. This effect increases with sparsity, dictated by total rather than active parameters. While 80M dense models can repeat data over 8x with minimal degradation, MoEs instead begin to suffer at 4x, and deteriorate rapidly, ceding their performance benefits in all-unique data settings to underperform dense models after 32x. We experiment with existing regularization methods as a potential remedy. We find that some methods, such as dropout, can mitigate overfitting. In particular, with strong masking-based regularization, MoEs are able to outperform dense models even when data is repeated more than 64 times. However, no method fully matches the performance of all-unique training data. Finally, we analyze internal mechanisms correlated with MoE overfitting in high repetition regimes, and find that MoE routing universally stabilizes early in training, and that expert specialization correlates with overfitting to repeated data. In sum, our work addresses the adverse interactions between sparsity and data repetition: we present evidence for the core mechanisms of overfitting and its potential remediation, and suggest promising avenues for future methods to reduce over-specialization in model parameters by disrupting memorization patterns.