Model Selection

Heterogeneous federated learning requires clients with diverse computational capacities to collaboratively train a global model, where each client trains a capacity-constrained submodel. Existing methods select submodel parameters using heuristic importance measures---most prominently parameter magnitude---without theoretical justification for why these measures support convergence. We identify a fundamental gap: existing parameter selection criteria lack theoretical grounding in the convergence framework, partial client participation introduces additional estimation effects in the Fisher scores. We propose \textbf{FedFIbOS}: Fisher Importance-based Optimal Submodelling for heterogeneous federated learning, using Fisher Information in a principled criterion derived from minimizing submodel masking error. %We formally establish when magnitude selection is equivalent to Fisher selection fail under non-IID heterogeneous federated learning. We theoretically formulate submodel selection through a Fisher-weighted quadratic masking surrogate and show that the raw Fisher top-$k$ rule implemented by FedFIbOS solves this surrogate under a Fisher-dominant ranking condition. The resulting method retains the convergence structure of the underlying masked federated optimization bound. Fisher scores are efficiently estimated from empirical diagonal Fisher information using squared gradients, enabling stable and adaptive parameter selection without additional optimization overhead. Experiments on CIFAR-10, CIFAR-100, and AGNews under pathological and Dirichlet non-IID settings show FedFIbOS achieves ${\approx}10\%$ higher accuracy than the state of the art, with improvements becoming more pronounced under stronger heterogeneity.

Classifiers can make identical predictions yet require labels to compare their selective performance: confidence ranks weight the same errors differently. We quantify this requirement for the area under the generalized risk-coverage curve (AUGRC). A prelabel lower bound rules out insufficient budgets. With all labels known, a covering linear program bounds the minimum number of labels sufficient to fix the winner (the certificate size) within $K-1$ labels for $K$ candidates. For fixed $K$, independent uniform orders and identical predictions, the prelabel bound approaches one quarter of the pool. With iid Bernoulli errors independent of the orders, every exact acquisition policy reads almost all labels asymptotically, although a two-candidate certificate needs only half. Across 108 feature-panel comparisons on nine datasets, disagreement labels settle every accuracy choice but no AUGRC choice. A 20% budget is ruled out in 96 conditions; certificates need 56-57% on average. On ten conditions with pretrained image classifiers, confidence-score choice reads 68-91% of 10,000 labels for exact selection and 50-67% with AUGRC tolerance $5\times10^{-4}$. An exact stopping test works with any acquisition order. Together, these results link confidence ranks to label budgets and certified model comparison.

Evolutionary prompt optimizers such as GEPA suffer from prompt bloat: each iteration appends rules and caveats, producing prompts up to 3$\times$ longer yet no more accurate. We trace this to three deficiencies - incomplete error observation, limited search diversity, and unreliable selection - and propose ESPO (Error-Structured Prompt Optimization), which decomposes prompt optimization into three phases: Diagnose clusters all training errors into structural patterns in one round; Propose generates candidates via four complementary strategies with independent biases; Select applies bootstrap stability selection. On seven public NLP benchmarks - Tweet, MMLU, GSM8K, HotpotQA, ScoNe, HoVer, and PUPA - ESPO improves average accuracy by $+$3.76 pp over the state-of-the-art (74.67% vs 70.91% for GEPA), matching or exceeding GEPA on every dataset while producing prompts 47% shorter (1,004 vs 1,878 chars) and faster at inference. Cross-model experiments across four additional student models (Gemma 3 12B, Mistral 14B, Qwen3 32B, Claude Haiku 4.5) show ESPO yields the best average accuracy on every model tested, with the largest gap on Qwen3 GSM8K (15.00% $\to$ 91.40%). A generalization bound (Appendix) grounds each phase in a corresponding term of the test-time gap, and the ablation confirms a key prediction: adding diversity without bootstrap selection actually hurts performance ($-$1.20%).

Model merging has recently attracted significant attention as a promising paradigm for constructing unified multi-task models without requiring additional retraining. However, parameter conflicts and knowledge interference across tasks often degrade merged-model performance. Prior work introduced Conflict-Aware and Balanced Sparsification (CABS), which reduces parameter interference through structured pruning and sequential masking. However, CABS relies on grid search to determine scaling coefficients, resulting in exponential time complexity, while its optimization objective can be dominated by high-performance tasks, leading to suboptimal overall performance. To address these limitations, we extend CABS and propose CABS+. Specifically, Adaptive Weight Allocation (AWA) optimizes merging coefficients via a gradient-free search scheme to reduce time complexity, while an asymmetric fitness function promotes more comprehensive performance gains across tasks. Moreover, we conduct a systematic empirical study of key factors influencing model merging performance and propose Relative Synergy Score (RSS) to quantify model mergeability and guide model selection. We compare CABS+ with state-of-the-art model merging methods, including CABS, AdaMerging, and WUDIMerging, across 27 datasets and 5 models covering large language, small-scale language, and vision models. Extensive experiments verify the effectiveness and efficiency of CABS+. Compared with AdaMerging and WUDIMerging, CABS+ improves overall performance by 16.97% and 12.93%, respectively, exhibits stronger stability and robustness across varying task numbers and model architectures, uses less than 25% of the GPU memory required by AdaMerging, and achieves nearly a 4x speedup in merging time over WUDIMerging.

Foundation models for protein structure prediction remain unreliable on certain targets. External oracles can flag and correct these failures, but biological oracles are expensive, making oracle budget a critical constraint. Existing guidance methods, such as FK-steering, DPO, and Best K-of-N sampling, differ in how they spend this budget, yet no systematic comparison exists to guide method selection. To bridge this gap, we benchmark these methods alongside the recently proposed Optimisation Over Outputs (O3), which applies off-the-shelf optimisers within a generative model's latent subspace. We extend the usage of O3 to protein structure prediction models. Overall, our work provides the first practical reference for oracle budget-aware guidance. Our evaluation on two protein targets, calmodulin (1CLL) and E. coli aspartate transcarbamoylase (9EEH), reveals that no single method consistently dominates across all budgets and oracles. Specifically, O3 proves most effective at low oracle budgets, while FK-steering and DPO demonstrate improved performance as the budget increases. We distil these findings into actionable recommendations for practitioners operating under real-world oracle-budget constraints.

Large multi-call LLM systems pose a scientific problem that query-level routing does not capture: the value of a model depends on where it enters a dependent computation and on the deployment that surrounds that call. Existing routers typically decide \emph{where} to spend a stronger model while treating the benefit of the substitution itself as known. We separate these two decisions through a predicate-action factorization and evaluate it in controlled solve-merge-verify workflows spanning 8-64 solve calls and four three-tier model ladders. The resulting evidence reveals a consistent principle beneath apparently conflicting outcomes. On numeric frequency counting, all-strong reduces RMSE from 4.818 to 1.538 in the Mixed Qwen/GPT ladder, whereas the average Qwen-only ordering reverses. Input-matched interventions further show that the same medium-to-strong action has sharply different value across roles and scales. A semantic task-and-contract shift reverses the Mixed ordering again, while allocation ablations distinguish useful sparse placement from under-coverage and indiscriminate escalation. Together, these results establish model substitution as a deployment-conditioned action rather than a property implied by a tier label, and they provide a practical sequence for large-scale workflow routing: calibrate the action, resolve its role-conditioned effect, and then optimize its placement.

Model selection for imbalanced binary classification often uses the Matthews correlation coefficient (MCC), but thresholding makes validation rankings threshold-dependent. SoftMCC is a post-training MCC validation framework on established probability-valued confusion counts, coupling an MCC-specific calibrated identity with a tie-aware, shared-pool selection protocol. Its core score is a covariance-normalized probability-label association, reduces exactly to MCC for hard predictions, and is Pearson-bounded. Under perfect population calibration it equals the Brier skill score with identical candidate ordering; outside that regime the gap does not identify calibration error. Across 18 settings with 12 duplicate-safe grouped repeats, SoftMCC attains the best stability mean rank (2.31) and highest mean tie-corrected Kendall's W (0.659), with a significant Friedman test (p=0.007); Nemenyi analysis separates it from AUPRC and MCC@0.5, while 14-source-family sensitivity retains only the latter. Selected-model utility shows no advantage. Three of six prespecified comparisons have negative mean test-MCC differences, only F1@best survives Holm correction (p=0.014), and the dataset-level test is not significant (p=0.117). Label permutation lowers mean W to 0.092; temperature scaling shifts SoftMCC rankings (mean Spearman 0.851) whereas rank-based and threshold-optimized metrics remain invariant. SoftMCC is a calibration-sensitive MCC-family selector with bounded stability and utility evidence.

Foundation models are increasingly reused as software components, making model selection a critical software-engineering decision. Current model hubs primarily support discovery through popularity metrics, often neglecting functional capabilities, operational constraints, and community-perceived quality. We argue that foundation-model selection should be treated as an explicit, auditable software-component selection task rather than as keyword search, popularity ranking, or opaque conversational advice. This paper proposes HugSelect, an explainable decision-support framework for foundation-model selection. HugSelect builds a knowledge base of 71,274 models by combining repository metadata, extracted functional capabilities, and perceived quality attributes derived from community discussions into a unified pipeline. It ranks candidate models using a weighted additive model that exposes criterion-level score decompositions. We evaluated HugSelect through pipeline validation, comparative case studies against four commercial LLM-based recommendation systems (44 scenarios), fine-grained ablation, and an exploratory user study (n = 10). Extraction pipelines achieved an F1 score of 0.801 for functional features and an accuracy of 0.84 for quality-attribute mapping. HugSelect achieved a model-level Coverage@10 of 0.61 and family-level Coverage@10 of 0.91, showing recommendation quality comparable to that of the evaluated commercial systems, with no significant overall differences in ranking quality, while providing stable, traceable, and inspectable reasoning. Ablation confirmed that functional features were the main driver of retrieval accuracy, and preliminary user feedback suggests that the framework is useful and intuitive.

Embodied intelligence demands both long-horizon reasoning and real-time closed-loop responsiveness. Recent dual-system Vision-Language-Action (VLA) architectures combine fast reactive control with slow deliberative reasoning to balance inference speed and task success rate. However, existing dual-process VLAs tightly couple the fast module to intermediate representations of the slow module, necessitating end-to-end joint training and limiting modularity, extensibility and flexible system switching. In this paper, we propose Environment-aware Model Selection (EMS), an adaptive VLA inference framework that switches between two fully decoupled systems of different scales through environment-aware model selection. The large-scale deliberative system provides globally consistent trajectory planning to ensure task success, while a lightweight reactive system enables high-frequency closed-loop control. A reinforcement-learning-based switching policy dynamically selects which system to invoke based on real-time feedback, enabling sparse use of the slow system and thereby balancing pretrained knowledge utilisation with runtime efficiency. Our design offers three key advantages over prior hierarchical VLA frameworks: (1) a fully decoupled and modular dual-system architecture that supports plug-and-play model replacement; (2) an adaptive, environment-aware switching strategy; (3) high-frequency inference for responsive closed-loop control. We extensively evaluate EMS in both simulation and real-world environments. On the LIBERO benchmark, EMS achieves success rates comparable to the large-scale baseline while increasing the effective action frequency to 93.4 Hz. The framework further demonstrates strong extensibility in real-world dual-arm manipulation tasks, where it accelerates task completion while maintaining robust performance.

Large language models (LLMs) are increasingly deployed in enterprise settings, yet individual models remain bounded by model-specific capability limitations. These heterogeneous boundaries pose a deployment challenge, but also create an opportunity: strategically coordinating multiple LLMs may unlock collective intelligence exceeding any single model. Existing approaches fix how models are combined in advance, overlooking the dynamic, state-dependent role of complementarity in complex problem solving. Drawing on the wisdom-of-crowds paradigm, we reconceptualize collective LLM intelligence as relay-style complementarity: a sequential process in which each successor model is selected to address the specific bottleneck identified in its predecessor's output. To operationalize this, we propose WILC (Wisdom Integration of LLM Crowds), a framework grounded in two design principles. First, iterative reflection-and-refinement establishes a state-preserving workflow through which models diagnose and refine prior outputs. Second, complementarity-driven model selection governs transitions via a dual-gate mechanism: prospective complementarity fit (PCF) identifies the worker most suited to the current bottleneck, while posterior complementarity gain (PCG) evaluates whether the selected transition improves the evolving solution. Experiments across four diverse benchmarks show that WILC outperforms existing approaches, including single-model self-refinement, ensemble methods, and query-routing methods. Under standardized pricing assumptions, WILC matches the average benchmark performance of GPT-5.2 at roughly 7 times lower estimated per-query cost, while facilitating data sovereignty through self-hosted deployment. This study extends wisdom-of-crowds theory from static aggregation to sequential AI complementarity and provides transferable design principles for multi-AI coordination.

Standard model comparison is global, aggregating losses across the covariate space to declare a single winner. This can obscure heterogeneous performance, where different models are preferable in different regions. We introduce conformalized local model comparison, a split-sample framework for constructing calibrated local best-model maps. Given a model comparison score, such as the difference between two squared losses, the method uses three disjoint splits to fit competing models, estimate local centers and scales from out-of-sample scores, and conformally calibrate residual uncertainty. At a target point, the procedure declares a local winner only when a one-sided conformal bound excludes a tie, with the score's sign determining the favored model. We prove finite-sample marginal control for one-sided erroneous declarations on the realized future comparison score, establish pointwise consistency of the localized mean-score estimator away from tie boundaries, show that aggregate comparison can disagree sharply with the prevalence of local superiority, and derive a squared-loss bias--variance decomposition that clarifies how model structure affects local wins. Synthetic and real-data experiments show that the method recovers heterogeneous winner regions, abstains under uncertainty, and yields higher conditional gain than global selection.

As decision-making processes grow more complex, machine learning tools have become essential for tackling business and societal challenges. However, many existing methods rely on decision-making procedures that are difficult to interpret. Since humans naturally make decisions by comparing new cases with a few representative examples, we aim to design an approach that selects such pivots to construct an interpretable predictive model. Inspired by decision trees, we propose a hierarchical, interpretable-by-design pivot selection model based on the similarity between pivots and input instances. Our method functions both as a pivot selection technique and a standalone predictive model. Extending beyond single pivots, we incorporate pairs of pivots that are used by proximity and oblique trees, as well as ensembles, which enhance the versatility and effectiveness of our proposal. Additionally, our approach is data modality-agnostic, leveraging pre-trained networks for data transformation. Experiments across diverse datasets, including tabular data, text, images, and time series, demonstrate the effectiveness of our approach, outperforming alternative instance selection strategies and achieving competitive results against state-of-the-art interpretable models while maintaining a minimal number of pivots.

Evaluating large generative models across benchmarks is time-consuming and computationally expensive. This drives the need for methods that can estimate full benchmark performance by evaluating models on only a subset of items, known as a coreset. Current literature mostly requires the practitioner to input a coreset size. However, when reliable performance estimation takes priority over efficiency, an evaluation method should also be capable of automatically determining a coreset size that reflects this priority. We introduce BayesAME, a sequential Bayesian framework specifically targeting automatic determination of the coreset size. BayesAME models performance as a random variable by defining a latent ability for each group of items sharing the same historical model performances, with a joint prior distribution encoding the belief that the target model behaves similarly to these historical models. The posterior distribution over these abilities is used to derive performance estimators, quantify performance uncertainty, and select items to add to the coreset via an information-gain criterion. The coreset is iteratively augmented until the performance estimate fluctuation and the performance uncertainty fall below their respective user-defined thresholds. We propose a multi-target extension that captures performance correlations across multiple target models to further reduce the coreset size. Through extensive experiments across diverse benchmarks, we demonstrate that BayesAME consistently outperforms sequential adaptations of existing methods. Crucially, our comprehensive analysis addresses recent skepticism in the literature, establishing that non-random coreset selection is advantageous over random selection. Finally, we highlight that leveraging continuous response log-likelihoods over traditional binary scores significantly enhances estimation accuracy.

Conformalized selection has been widely applied to select high-quality candidates from large datasets with rigorous uncertainty quantification, such as reliable labeling, drug discovery, and the alignment of large language models. Nevertheless, existing methods assume clean responses on calibration data, an assumption that rarely holds in practice. In this paper, we formulate the above tasks as selecting candidates with true predicted labels or with responses exceeding certain values. We demonstrate that existing conformal selection methods fail to control the false discovery rate (FDR) or suffer from severe power loss under contaminated calibration data. To that end, we propose Robust Conformalized Selection (RCS), a unified framework for selective classification with valid FDR control under general label contamination. The key insight of RCS lies in a novel statistical reduction: by separately conditioning on different classes, we translate the intractable label noise into a localized covariate shift problem, which then enables a covariate-adjusted empirical-Bayes-type estimate of the number of false selections. Statistical properties such as the asymptotic FDR control, power optimality, and robustness of RCS are established. We further develop an instantiation of RCS under randomized response model, and also apply RCS to the task of selecting candidates with large response values. Extensive experiments on both simulated and real-world datasets demonstrate the effectiveness of RCS.

Closed-loop generative selection has become a workhorse of computational drug discovery: a learned generative model proposes candidate molecules, a fitness oracle scores them, the best are kept, and the model is retrained on this elite set before the next round. Despite its wide use, the method has lacked a rigorous convergence theory, largely because retraining the model each round breaks the Markov property on which classical evolutionary-algorithm analysis relies. We develop a self-contained theory of convergence and expected running time for this class of algorithms. By recovering a Markov structure on an enlarged state space, we show that elitism makes the search absorbing, and we prove almost-sure convergence together with a runtime bound that decomposes the search into the time spent escaping each fitness level. We then analyse the role of the model's memory---how much of the past it is trained on. When learning improves steadily with more data, deeper memory never hurts; when it does not, an exit-time analysis pinpoints the optimal memory depth and shows that excess memory can actually slow convergence. The theory extends to multi-objective search and to noisy oracles: we quantify how many repeated evaluations certify progress under light-tailed noise, and how robust estimators restore guarantees under heavy tails. Recast in terms of oracle evaluations - the true bottleneck in drug design - the analysis yields a concrete, evaluation-minimal strategy. Areproducible study confirms the predictions, including the surprising cost of excess memory. We close with three open problems.

B-spline regression constitutes a widely used framework for nonparametric modeling. The performance of this methodology depends on specifying the number and placement of changepoints, known as knots, prior to the estimation process. Such knot sequence determines the dimension of the B-spline basis used to represent the regression function and the number of coefficients to be estimated. Therefore, the knots' choice affects the model's flexibility, influencing its smoothness and goodness-of-fit. Traditionally, this problem has been addressed either by explicitly selecting knots, via knot-selection algorithms, or by regularization methods, such as P-splines, which automatically tune the regressor's smoothness. The latter have become the standard in generalized additive models (GAMs). In contrast, knot-selection techniques, frequently neglected because of computational or modeling limitations, provide certain advantages which can be valuable in some contexts. In this work, we introduce a novel explicit knot-selection technique for GAMs based on an extension of the adaptive splines (A-splines) knot selection methodology, combined with a customized Fellner-Schall scheme for tuning the associated parameters. Our approach is evaluated on various synthetic and real datasets and compared with P-splines and state-of-the-art knot-selection techniques. The results indicate comparable performance, while producing models built on a substantially smaller number of basis elements.

Neural network compression and interpretability remain open challenges in modern deep learn- ing, where billion-parameter architectures deliver impressive accuracy at the cost of trans- parency, computational efficiency, and reliable uncertainty quantification. This paper introduces Neural Atom Prevalence (NAP), a principled Bayesian framework for structured node-level model selection in feedforward neural networks. NAP introduces the neural atom (activation unit) and functions as a hybrid method operating through a four-phase pipeline: Bayesian Lottery Ticket (BLT) identification via Iterative Magnitude Pruning (IMP), soft variational training of the Spike and Slab Independent Gaussian (SS-IG) model, Poisson-Binomial (PB) optimal layer-size selection, and Bayesian fine-tuning to produce a sparse, stable, interpretable, and accurate model. Extensive empirical validation across simulated nonlinear regression, two UCI benchmark datasets (Concrete, YearPredictionMSD), and the MNIST image classification task demonstrates that NAP achieves state-of-the-art structural sparsity, reducing active nodes to as few as 8% of the original dense architecture on MNIST, while well-calibrated probabilisti- cally: the aleatoric-epistemic uncertainty decomposition reveals that model ignorance accounts for only 3 to 4% of total predictive variance across all experiments, and regression reliability diagrams confirm a near-nominal predictive interval coverage (93.4% observed against a 95% target). These results establish NAP as a reliable, theoretically grounded, and computation- ally tractable solution to the simultaneous pursuit of sparsity, accuracy, interpretability, and uncertainty quantification in Bayesian neural networks.

Video capsule endoscopy (VCE) classification is typically evaluated within a single dataset, yet clinical deployment demands robustness across acquisition sources, labeling policies, and patient populations. We examine this gap using Kvasir-Capsule, Capsule Vision 2024 (CV2024), and a shared-label subset of Galar. We fine-tune a suite of general-domain pretrained backbones on the official Kvasir-Capsule folds under a standardized protocol and evaluate the same checkpoints on two non-source targets within a documented shared-label decision space. We find that the predictive value of in-domain ranking is target-dependent: Kvasir-Capsule ranking aligns more closely with Galar than with CV2024, while the two non-source targets agree only weakly. Consequently, the strongest in-domain backbone leads on one target yet falls to mid-pack on the other, and no single evaluation target reliably predicts the others. A second CV2024-trained configuration set reproduces this target-dependent instability. We conclude that capsule endoscopy model selection should report cross-target ranking stability rather than peak single-dataset performance.

Cross-Project Defect Prediction (CPDP) involves building models using data from external projects, called training projects, to predict modules from the target project. However, traditional CPDP methods suffer from the distribution shift between training and target projects that affects the model's performance. This paper proposes a novel CPDP framework that addresses this issue by proposing a two-stage multiple classifier system (MCS) selection scheme: one working at the project level and another at the module level. In the first stage, the framework evaluates multiple possible MCS configurations to find one that covers and generalizes well across multiple training projects. Consequently, the proposal is likely to obtain a diverse set of classifiers, each specialized in tackling software modules with distinct characteristics. The second selection stage operates at test time, selecting the most competent classifiers to predict each new module in the target project. Unlike previous approaches that apply the same classifiers to the entire target project, the proposed framework performs module-level model selection. This way, the system is more robust to changes in distributions between training and target projects because the selected set of classifiers is module-dependent. Our experimental results using 82 projects from four different CPDP benchmark datasets demonstrate that the proposed approach outperforms the state-of-the-art CPDP methods in most scenarios. The code, dataset, and further details about the proposed method are publicly available at https://github.com/jsaj/Multi_DES.

Neural simulation-based inference enables parameter estimation for complex models, but typically requires the user to specify a simulator encoding a fixed model structure. We present a framework for joint model selection and parameter estimation that combines large language models for program synthesis with neural simulation-based inference. Given a natural language description of the system and data under investigation, an LLM proposes candidate simulator programs which are iteratively refined via feedback-driven mutation and evaluated using neural density estimation. The approach enables simulation-based inference over a pool of models, not just parameters within a fixed model. On benchmarks spanning deterministic dynamics, stochastic epidemic models, and dark matter substructure inference from gravitational-lensing images, the method identifies plausible model families from open-ended prompts, with accuracy that reflects the information content of the data and identifiability of candidate models.

We introduce Deep Adaptive Bayesian Screening (DABS), a method for performing adaptive factorial screening in high-dimensional discrete design spaces. DABS learns a policy network offline to sequentially select informative experiments, amortizing Bayesian Optimal Experimental Design. It handles binary designs, incorporates sparsity and interactions via a spike-and-slab prior with strong heredity. The model is trained using a contrastive lower bound on information about factor activity with nuisance effect sizes and noise variance analytically integrated out. Unlike prior amortized Bayesian design approaches, DABS also integrates Gibbs posterior inference at deployment, yielding posterior probabilities of factor activity and credible intervals on effect sizes. We demonstrate DABS on screening problems calibrated to real-world benchmarks and show it achieves superior accuracy and scalability over classical and Bayesian baselines under tight experimental budgets.

Selective state-space models such as Mamba route information through a bank of first-order modes whose input coupling is set by a learned selection mechanism. We give an exact instrument for measuring how a trained model uses these modes. Because the state matrix is diagonal, each channel's output decomposes exactly into per-mode contributions, and a per-(layer, channel, window) Gram tensor yields the exact output error of dropping any subset of modes, offline, at any budget. Validated against the reference implementation to a relative error of $2.3\times10^{-7}$ on the Mamba-1 family where it is exact, the instrument predicts a layer's deployed pruning error to a median relative deviation of $5\times10^{-7}$ over $4{,}464$ configurations, its floor set by the reconstruction. Applying the instrument across the Mamba-1 family (130M--2.8B), the deployed 7B Falcon-Mamba, and Mamba-2, we find that trained models re-allocate their state space with the input: which modes carry the signal migrates across contexts, and at the most affected layers a per-input oracle roughly halves the output error of a fixed mode set. Frozen-signal counterfactuals attribute the migration primarily to the input-dependent write map $B_t$; the timestep usually identified with selectivity carries almost none of it. Input-scheduled mode pruning on this measurement outperforms static, Hankel-based, and layer-adaptive rankings at every scale from 130M to the deployed 7B Falcon-Mamba, and at half the state budget it matches the unpruned model. Because the scheduler reads each window's mode usage from a first pass, this demonstrates realizable headroom; we claim no deployed compute or memory saving.

Numerical integration is a cornerstone of various scientific computing applications, such as engineering simulations and model evidence computations in probabilistic machine learning. Bayesian Quadrature uses Gaussian process surrogates that explicitly encode structural assumptions about the integrand to obtain integral estimates with quantified uncertainty. These surrogates are predominantly based on stationary covariance functions, which results in model misspecification for integrands exhibiting nonstationary behavior. We tackle this issue through an adaptively growing, tree-based partition of the integration domain into local stationary models. Our method recombines the local integral estimates through a hierarchy of GP conditioning that reintroduces cross-subdomain correlations, while model selection criteria control the tree growth to avoid unnecessary partitioning. The resulting algorithm is simple, requires no MCMC, and adapts its evaluation budget to local integrand complexity. On benchmark integration problems and a model evidence computation for an epidemiological model, Hierarchical Bayesian Quadrature achieves substantial gains over standard Bayesian Quadrature on nonstationary integrands while matching its performance on stationary ones.

After a weak verifier accepts a large-language-model response, a second call may resample or reroute. Because correctness is hidden, action selection is an identification problem. We order three gates: recoverable stopping debt, two-sided FIT action support, and held-out value from an outcome-blind selector. In a pinned 152-query MBPP+ experiment, a Qwen2.5-14B Base-only false-positive stop leaves +2.592 percentage points of Qwen2.5-7B recovery (query-cluster 95% interval [+1.618, +3.664]). Separately, after 7B Base-test rejection, fixed escalation to 14B exceeds leave-one-out 7B resampling by +2.882 points [+0.931, +5.201]; this is fixed-action ranking, not conditional selection. An all-episode audit produces a +2.697-point realized-maximum gap, but for two actions this statistic equals (1/2)E|Delta| - (1/2)|E Delta| and contains no observable-history term. It lies inside an exact-fold exchangeable reference (mean +3.158; 95% interval [+2.434, +3.947]). The audit unconditionally acts on 1,520 episodes: 1,240 observable stops and 280 verifier rejections; 198 stops are evaluator-only false positives. Neither tested outcome-blind controller improves on fixed rerouting. A separate LiveCodeBench ladder has all-zero FIT action advantages despite exclusive TEST rescues. A preregistered BigCodeBench support gate then finds only 23/19 and 22/19 signed episodes/queries against minima of 25/20, so L1-L4, DEV, and TEST stay unopened. Stopping debt exists, but current evidence does not identify when to resample rather than reroute.

In partially exploratory factor analysis (PEFA), the loading structure and factor numbers are weakly specified. The regularized variational approximation for partially confirmatory factor analysis (PCFA VA) recovers this structure via Bayesian variable selection, using spike and slab priors to assign inclusion probabilities to unspecified loadings. This research introduces a post selection assessment framework for this approach. We convert converged solutions into covariance models using either hard selection (thresholding probabilities into a sparse pattern) or soft selection (retaining them as weights for effective parameter counts). We derive the resulting degrees of freedom, absolute fit diagnostics (RMSEA, SRMR, CFI, TLI), and relative criteria (AIC, BIC, ELBO). To determine factor numbers, we propose a scale free gain rule with a sustained drop guard. Simulations show absolute indices successfully track loading recovery and flag under factoring. While raw criteria over factor, our gain rule accurately recovers true dimensionality, with the ELBO variant proving most robust. Finally, a 100 item PID 5 example demonstrates that our model fits better than a confirmatory 25 facet model and concordantly recovers major structures across disjoint specifications.

Coreset selection aims to identify a small and highly representative subset of a massive dataset for efficient model training. The problem remains challenging even in the few-shot knowledge distillation (KD) setup, where a full-scale pre-trained teacher informs the student network. Typical sample selection strategies often struggle to surpass the random selection baseline. In this paper, we showcase few-medoids, an embarrassingly simple coreset selection strategy that chooses the samples closest to the centroid (average image) of each class. We present extensive KD experiments on four datasets, covering a wide range of image classification problems, and three teacher-student model pairs, comprising both convolutional and transformer networks. Although the proposed method is embarrassingly simple, our empirical results indicate that few-medoids is able to consistently surpass the random selection baseline, as well as the other coreset selection strategies. We therefore consider that few-medoids can be used as a drop-in replacement for commonly-used baselines (e.g. herding or k-center Greedy), in future research on coreset selection. To reproduce the reported results, we publicly release our code at https://github.com/CemilAndreiDilmac/Few-Shot-KD-Coreset.

Teams deploying large language models in business contexts need evaluation systems, yet most treat evaluation as static model selection: run benchmarks, rank models, deploy the winner. This framing misses evaluation's primary value for production systems--diagnosing why a system underperforms and guiding what to fix. We present EvalLoop, a methodology for evaluation-driven iterative improvement. EvalLoop organizes evaluation around three mechanisms: (1) dimensional metric grouping that decomposes quality into business-relevant dimensions enabling orthogonal failure diagnosis; (2) failure mode classification that categorizes why outputs fail within weak dimensions, bridging diagnosis to action; and (3) a structured iteration workflow where each evaluation run varies one system variable and compares dimensional profiles before and after. We validate EvalLoop through a case study on sales intelligence briefing generation (10 models, 3 providers, 18 metrics, 5 dimensions, 3 iterations). Dimensional diagnosis identified that 69% of hallucination failures were prompt-induced interpretation errors--invisible in aggregate scoring. A targeted prompt fix improved the best model from 82.6% to 94.6% overall, with improvement concentrated in diagnosed dimensions (Content Accuracy +16.8pp, Synthesis Power +26.4pp). An undirected configuration change in a prior iteration produced zero impact, illustrating the cost of iterating without diagnosis. We additionally demonstrate that dimensional profiling enables deployment-specific model selection, and that a one-time blind human gate on a finalist panel (4 models, 16 cases) confirms dimensional rankings while resolving multi-criteria deployment trade-offs--a 94% reduction in review burden compared to evaluating the full design. EvalLoop is packaged as reusable artifacts (playbook, agent specification, template repository) for adoption by other teams.

First-shot anomalous sound detection in DCASE Challenge Task 2 must flag anomalies of unseen machine types with a single threshold, without knowing whether a test clip comes from the data-rich source domain (990 normal training clips) or the data-scarce target domain (10). Two organizer-reported problems remain open: source- and target-domain AUC are negatively correlated across systems, and development-set performance does not predict evaluation-set performance. We address both with a training-free post-hoc layer over frozen audio embeddings: (i) per-domain quantile calibration shrunk toward a pooled map by a prior strength m, tracing a source/target balance frontier, and (ii) a label-free cross-validated domain-balance criterion that ranks candidate configurations from training normals only, paired with a coarse development-labeled viability veto. On DCASE 2025, the criterion rank-predicts the official evaluation score across a 45-configuration grid (Spearman rho = +0.91; family-block bootstrap 95% CI [+0.83, +0.95]) while development score is uninformative (+0.06). Criterion-based selection raises the evaluation score from 55.83 to 59.34 (jackknife CI [2.2, 4.8]) and, on an extended grid, to 61.05 -- retrospectively fourth of 35 teams. Replicating on DCASE 2023 and 2024 bounds the claim: development score is uninformative in all three years and degenerate configurations recur (vetoed every time), but under family-clustered uncertainty the criterion's predictive evidence survives only in 2025; in both replication years a fixed full-equalization default matches or beats criterion-based selection. A DCASE 2026 forward test is frozen before the 2026 evaluation ground truth is released; all headline numbers are reproduced by the official evaluator.

The aggregation with exponential weights (AEW) estimator is not fully understood in the basic setting of model selection aggregation with squared loss. In particular, whether it is minimax-rate optimal in expectation for large enough fixed temperatures and under random design has been an open problem since its introduction, which was explicitly posed by Lecué and Mendelson (2013). In this paper, we settle this problem by showing that \emph{without} requiring a Bernstein-type assumption, the AEW indeed achieves the excess risk $T \log (M) / (n+1)$ in expectation, whenever the temperature $T$ satisfies $(L^2/T)\exp(B/T)\leq μ/2$. Here, the number of dictionary elements is $M$, the estimator has observed $n$ i.i.d. samples from any distribution, and the loss is assumed to be bounded by $B$, $L$-Lipschitz continuous and $μ$-strongly convex. For squared loss, we show that $T\geq 4 b^2$ suffices when the predictions and labels are $[0,b]$-valued. Because AEW is known to be suboptimal in expectation for temperatures below some constant, this shows that AEW has a sharp phase transition when the temperature is large enough but constant, as conjectured by Lecué and Mendelson.

Ensuring model reliability in Explainable AI requires a global assessment of the hypothesis space. We propose a formal framework for the exhaustive analysis of optimal and near-optimal decision trees, called Algebraic Decision Tree Counting (ADTC). Inspired by Algebraic Model Counting (AMC) in knowledge representation, ADTC reformulates diverse analytical tasks, such as optimization, counting, and sampling, into a unified sum-of-products computation over a semiring $R$. While the hypothesis space of decision trees is doubly exponential with respect to the maximum depth $Δ$, our dynamic programming algorithm achieves $O^*(n^{O(Δ)})$ time complexity in the number of features $n$, where $O^*$ suppresses polynomial factors. To handle complex constraints consisting of multiple tree metrics, we introduce model behavior tensors that aggregate semiring values via convolution products over a tensor semiring. This algebraic approach efficiently constructs a model profile that captures the global landscape and trade-offs between criteria such as accuracy, size, and fairness. We demonstrate the utility of our software, emtrees, on real-world datasets, illustrating how ADTC facilitates evidence-based model selection in sensitive domains.

A central challenge in statistical modeling is identifying the subset of features that belong in the true regression model. The classical best subset selection problem, recently made tractable via mixed-integer optimization (MIO), finds the globally optimal sparse solution. It does not, however, make use of any information beyond the observed data. In many applied settings, domain experts can meaningfully rank or score the relevance of candidate predictors, yet no existing framework integrates such probabilistic expert assessments directly into the best-subsets objective. This paper presents Expert-Implied Bayesian Best Subsets (EBBS), a method that incorporates domain-expert probability estimates of feature relevance into the MIO best-subsets problem through a maximum a posteriori (MAP) framework. Expert views from multiple respondents are aggregated into a single prior probability per feature using the Poisson binomial distribution for marginal probability estimates, the pairwise win rate for pairwise comparisons, or the normalized mean rank for ordinal rankings. This probability enters the objective function as a log-odds penalty term that smoothly encourages or discourages the selection of each feature consistent with the expert consensus. This paper provides analytic derivations of the MAP formulation and characterizes its theoretical properties. The proposed model reduces to Best Subsets when experts all have no views. Empirical results on synthetic and real datasets are forthcoming.

We study the problem of model selection among probabilistic forecasting models evaluated on datasets of multiple time series. The performance of a model on a single time series is quantified by the average value (score) of a proper scoring rule over a test set, but extending model selection to data sets of time series requires aggregating these scores. Common approaches either rely on scaling scores and averaging them (mean scaled score) or avoid scaling by using alternative statistics such as mean ranks or win rates. However, these approaches can yield conflicting conclusions. We show that such discrepancies arise from the skewness of the distribution of the scores, which is particularly pronounced when test sets are short. The skewness can cause non-mean criteria (e.g., mean rank, median, win rate) to select misspecified models. In contrast, the mean score is immune from this problem. We further show that, as the size of the test sets increases, all aggregation criteria converge to the same model selection decision, mitigating these discrepancies. Our experiments on intermittent demand time series, including data from the M5 competition, highlight the importance of sufficiently large test sets; the mean scaled score appears to be the more reliable approach, also because empirically we found its decision to remain consistent when different scaling factors are adopted.

While much effort has focused on developing and benchmarking high-performance neural networks, less attention has been given to how dataset properties, known to practitioners, can guide efficient model selection. Neural models are typically evaluated on datasets with thousands of classes, yet many real-world applications involve fewer than ten. To address this understudied but common setting, we develop a measure of classification difficulty based on data-side properties and show how it enables more efficient model selection for few-class datasets, where traditional approaches are less effective. We term this phenomenon "few-class distinctiveness". Our metric allows comparison of models and datasets 6 to 29$\times$ faster than repeated training and testing. Leveraging this insight, we extend scaled model families below the smallest published models, achieving greater efficiency at similar accuracy, for example models up to 42% smaller than YOLOv5-nano for a mobile robot task. Targeting resource-constrained applications, we demonstrate few-class model selection across mobile robot, drone, and IoT scenarios, highlighting practical gains in efficiency without sacrificing performance.

Distinguishing causal adverse drug events (ADEs) from spurious correlations remains a central challenge in pharmacovigilance. The InferBERT framework integrates transformer models with Do-calculus, but its success hinges on the underlying classification model. This study evaluates the impact of model choice in InferBERT, assessing whether simpler models suffice, if domain-specific pre-training helps, whether scaling to LLMs improves causal detection, and the effect of post-hoc calibration. We performed a comparative study on two benchmarks: Analgesics-induced Acute Liver Failure (AILF) and Tramadol-related Mortalities (TRAM). Four models were evaluated-XGBoost (baseline), ALBERT (original InferBERT), BioBERT (biomedical transformer), and Med-LLaMA (medical LLM)-using 5-fold cross-validation repeated over 20 runs. We measured accuracy, Expected Calibration Error (ECE) pre- and post-isotonic regression, and Jaccard concordance of causal terms with PRR, ROR, and EBGM; significance was tested with paired t-tests. BioBERT achieved the highest accuracy on both datasets, while Med-LLaMA underperformed despite its size and parameter-efficient fine-tuning. Domain-specific pre-training was decisive. Calibration improved ECE but had mixed effects on accuracy and causal discovery. BioBERT's superiority also yielded the strongest concordance with traditional pharmacovigilance signals. These results show that domain-specific pre-training provides a clear advantage over simpler baselines and larger LLMs. Investing in manageable, domain-aware models is more effective for computational pharmacovigilance than simply scaling model size.

Applications in the 3D Computing Continuum, which unifies edge, cloud, and space, require combining multiple AI tasks such as object detection, time-series analytics, and natural language processing into Compound AI systems. These systems must satisfy stringent Service Level Objectives (SLOs) on accuracy, latency, and cost. A key mechanism for maintaining SLO compliance of Compound AI systems is runtime model selection, where AI models are dynamically switched for each workflow task. However, existing distributed and compound AI frameworks do not natively support runtime model selection. We present PLAIground, a framework that enables runtime model selection for Compound AI systems. PLAIground introduces Compoundable AI Model (CAIM) abstraction, which decouples task semantics from AI model implementations via Task and Data Contracts, enabling model switching without workflow changes. Additionally, PLAIground introduces Pixie, an SLO-driven runtime model selection algorithm, which dynamically selects the most suitable model for each task during execution. Our evaluation on two realistic Compound AI workflows demonstrates that Pixie achieves up to 91.3% accuracy while maintaining SLO compliance where fixed-model strategies either violate cost and latency budgets up to 21x or miss accuracy targets by 4%.

The proliferation of recursive training on synthetic data can alleviate data scarcity but risks model collapse, where repeated training erodes distributional tails and homogenizes outputs. Data selection is widely viewed as a remedy, yet its reliability depends critically on the reference distribution used by the verifier. We show that in low-resource verification regimes, where each verifier observes only a small, fragmented, and biased slice of the target manifold, selection itself becomes biased. This situation naturally arises in low-resource data silos such as healthcare consortia or proprietary financial institutions, where raw data cannot be pooled and local references are inherently incomplete. As a result, selection preferentially retains samples aligned with the local manifold while pruning globally relevant tail modes, turning from a safeguard against collapse into a mechanism that precipitates it. We theoretically prove that such siloed selection accelerates collapse and induces power-law diversity decay. As an initial mitigation, we construct Wasserstein proxy references from multiple silos without sharing raw data. Empirical results confirm that local-reference selection fails on skewed distributions, whereas collaborative proxy references mitigate diversity degradation, suggesting that recursive synthetic-data pipelines require particular caution when real-data coverage is fragmented or scarce.

A central principle in quantum machine learning is that an ansatz should be expressive enough to represent the quantum data of interest. Yet, the expressibility is statistically meaningful only insofar as it can be learned from finitely many copies of an unknown quantum state. In this work, we develop an information-theoretic Occam theory for quantum data generated by finite-size quantum circuits. For the class $S_{n,G}$ of $n$-qubit pure states preparable with at most $G$ two-qubit gates, a metric-entropy argument gives the realizable sample law $\widetildeΘ(G/ε^2)$ in the circuit-limited regime. For an arbitrary source $\hatρ$, we introduce the best $G$-gate approximation error $d_G(\hatρ)$ and the approximate circuit complexity $C_η(\hatρ)$. We prove an agnostic quantum Occam theorem: with $M$ copies, one can learn up to the best $G$-gate approximation error plus a statistical penalty $\widetilde{O}(\sqrt{G/M})$. We then remove the need to know $G$ in advance through an adaptive model-selection theorem whose oracle inequality selects the circuit complexity justified by the data. Matching lower bounds yield a sample-supported expressibility law: at trace-distance accuracy $ε$, $M$ samples can support only $G_{\rm supported} \simeq Mε^2$ gates, up to logarithmic factors and tomography saturation at $2^n$. Thus, the circuit complexity becomes an adaptive statistical resource rather than a static promise. Our framework turns bounded circuit complexity into a model-selection principle for quantum machine learning.

Speech foundation models often struggle in low-resource domains due to domain mismatch and data scarcity. We propose Gumbel-BEARD, a domain adaptation framework that automates Whisper encoder layer selection via an end-to-end trainable hard Gumbel-Softmax selector. It enables self-supervised adaptation with a BEST-RQ objective that dynamically adapts to target acoustic characteristics without manual tuning. Experiments on the MyST child speech corpus demonstrate efficiency and scalability: with 10 h of labeled data for fine-tuning, our method matches a fully supervised baseline trained on the complete 133 h labeled set. We establish new state-of-the-art word error rates (WERs) of 8.21% using Whisper-medium on MyST and 11.06% using Whisper-small on the OGI Spontaneous dataset. Evaluation on CORAAL further confirms robustness to adult dialectal domain shifts, with up to 6% relative WER reduction, highlighting the generalizability of our approach to diverse low-resource conditions.

Abnormality detection is a crucial yet challenging task in medical image analysis. Distinguishing abnormalities from normal data by learning to reconstruct normal-only data alleviates the reliance on labeled datasets. However, many studies, even if unsupervised, rely on a labeled validation set to select the best model for inference from multiple training iterations. For many diseases labeled data are unavailable and substantially time consuming to obtain. To address this, AUCp - a novel metric that supports abnormality detection for unsupervised and self-supervised methods is proposed. Instead of evaluating the realism of reconstructed images to select the best of model for inference, it focuses on actual detection performance and without requiring an annotated test set. Assuming the pseudo ground truth of all unannotated samples in the test set as abnormal/positive and using traditional AUC calculation, AUCp scores are derived. Given a large and representative training set of normal samples, we show mathematical and empirical evidence that model selection using AUCp scores improves disease detection in terms of unsupervised and self-supervised methods over conventional metrics. Using two unsupervised methods for neurologic disease detection and self-supervised methods on diverse datasets, our results demonstrate that the AUCp score effectively identifies the optimal model for inference, significantly enhancing abnormality and disease detection. The corresponding implementations are available in https://github.com/mahfuzmohammad/AUCp.

Finetuning data selection requires balancing two competing goals: selecting examples that improve the downstream objective, and doing so without repeatedly finetuning models. Train-free selectors are scalable but rely on proxies such as embedding similarity or clustering, which may not match the target objective. Train-based selectors better reflect downstream utility through gradient signals, subset evaluation, or Shapley attribution, but require many costly train--evaluate iterations. We propose Hierarchical Active Region Pruning (HARP), an efficient train-based selector that preserves downstream alignment while reducing selection cost. HARP organizes the training pool into a node--leaf hierarchy, evaluates only representative leaves, and infers unmeasured utilities with empirical Bayes posteriors. It then selects data using two complementary envelopes: HARP-C, which conservatively controls redundancy, and HARP-E, which additively rewards complementary regions. We theoretically show that, under local smoothness and bounded estimation error, HARP controls selection error while reducing train--evaluate cost. We further validate that HARP variants achieve the best result and outperform the strongest baseline by up to $+8.9$ points, while using roughly $7\times$ fewer training examples.

TorchKM is an open-source library for kernel machines, including support vector machines, kernel logistic regression, and kernel quantile regression, with GPU acceleration. The library features a scikit-learn-style API and is designed to exploit GPU-friendly linear algebra, accelerating the full training and model-selection pipeline through intelligent reuse of matrix operations. Benchmarks show competitive predictive performance with substantial speedups over standard baselines. The efficiency and programmable design also make TorchKM a kernel-learning component for AI-driven workflows. Code and documentation are available at https://github.com/YikaiZhang95/torchkm, and the package can be easily installed via PyPI.

Understanding potential selection in data is crucial for causal discovery; we argue that "selection" in common narratives takes two forms, which we term static and evolutionary selection, respectively. Static selection refers to a one-shot filtering process where observed data consist of a subset of the population of interest, as in survey volunteer bias. Evolutionary selection, in contrast, operates through repeated rounds of differential fitness in reproduction, where observed data constitute the latest generation shaped by a historical trajectory, as in immune adaptation, antibiotic resistance, and social norm emergence. Existing methods largely conflate these two forms and rely on an identical graphical model of selection. We show that this model is valid for static settings but fails to characterize data under evolution, yielding false discovery results. To address this, we introduce a new model that specifically characterizes evolutionary selection, and develop a sound and complete procedure for identifying such models from data across one or multiple environments or generations. Experimental results validate the method's ability to uncover the relevant mechanisms underlying evolution from data.

Telehealth systems have become increasingly important for delivering accessible and timely medical information. Existing large language models often struggle to provide consistent and contextually appropriate medical responses across varying levels of case severity. This limitation highlights the need for models that can effectively adapt to the progressive complexity in medical queries. To address this challenge, we introduce a severity-aware multi-model framework that integrates curriculum training strategy with relevance-based response selection. The proposed framework employs a three-stage curriculum learning strategy, where each model is trained sequentially on mild, moderate, and critical cases to progressively acquire domain knowledge. The approach uses five large language models, each trained independently under the same curriculum. During inference, all models generate candidate responses, and the response with highest BERTScore is selected as the final output. The framework is trained and evaluated on the MAQA dataset, which provides annotated medical question-answer pairs. Experimental results evaluated using BERTScore demonstrate that the proposed method achieves superior performance compared to both baseline and fine-tuned models, attaining 86.71% in the baseline setting and 90.30% after fine-tuning. These results highlight the effectiveness of combining curriculum learning with multi-model response selection in improving response quality and relevance in medical text generation.

Automatic text summarization has become increasingly important due to the rapid growth of digital textual information. This paper presents a Multi-Model Summarization Framework designed to improve the robustness and quality of abstractive text summarization. Relying on a single model often leads to inconsistent summarization quality across articles with varying structures and topics. To address this limitation, the proposed framework integrates multiple fine-tuned transformer-based summarization models and introduces a metric-based selection mechanism. In this framework, each model independently generates a candidate summary for the same input article. The generated summaries are then evaluated using automatic evaluation metrics that capture both lexical similarity and semantic relevance. Based on these scores, the framework selects the highest-quality summary as the final output. The models are fine-tuned and evaluated on the widely used CNN/DailyMail news summarization dataset. Experimental results demonstrate that the proposed framework achieves the highest BERTScore among all compared methods with a score of 88.63%. It also outperforms several LLMs such as GPT3-D2, Falcon-7b, and Mpt-7b, highlighting its effectiveness and robustness. These findings highlight the effectiveness of leveraging multiple transformer-based models within a metric-based selection strategy to improve the quality and robustness of automatic text summarization systems.

Deep unsupervised domain adaptation (Deep UDA) methods successfully leverage rich labeled data in a source domain to boost the performance on related but unlabeled data in a target domain. However, algorithm comparison is cumbersome in Deep UDA due to the absence of accurate and standardized model selection method, posing an obstacle to further advances in the field. Existing model selection methods for Deep UDA are either highly biased, restricted, unstable, or even controversial (requiring labeled target data). To this end, we propose \textit{Deep Embedded Validation} (\textbf{DEV}), which embeds adapted feature representation into the validation procedure to obtain unbiased estimation of the target risk with bounded variance. The variance is further reduced by the technique of control variate. The efficacy of the method has been justified both theoretically and empirically.

Multi-step time series forecasting (MSF) is commonly evaluated using point-wise error metrics such as mean squared error (MSE), implicitly treating the conditional mean as a sufficient target. We show that this can be misleading under conditional uncertainty, where the conditional expectation becomes unrepresentative of typical realized values at longer horizons. We formalize this effect through a conditional uncertainty gap and prove that whenever this gap is nonzero, no deterministic predictor can simultaneously minimize MSE and match the marginal distribution of realized futures. This establishes a fundamental, model-agnostic trade-off between point accuracy and marginal realism in MSF evaluation. Using controlled stochastic dynamical systems and nine real-world forecasting benchmarks, we empirically characterize the resulting accuracy--realism frontier and \textbf{quantify the practical cost of MSE-only model selection}. As conditional uncertainty increases with forecast horizon, the attainable set expands into a pronounced Pareto front, separating MSE-optimal but under-dispersed predictors from methods that trade accuracy for realistic marginal variability. \textbf{Across benchmarks, we find that small relaxations in MSE ($\boldsymbol{\le 5\%}$) frequently unlock disproportionate gains in marginal realism, with median improvements of $\mathbf{17.3\%}$ and gains exceeding $\mathbf{30\%}$ in some datasets.} We further show that common forecasting strategies systematically occupy different regions of this frontier: direct multi-output predictors concentrate near the accuracy-optimal extreme, while recursive strategies and sample-based inference favors marginal realism. Together, these results expose a structural failure mode of MSE-based evaluation in long-horizon forecasting and recast strategy and inference selection as navigation of an unavoidable accuracy--realism trade-off.

Different predictors often excel on different inputs, so picking the best one per instance promises higher accuracy than committing to a single model. In practice, selectors trained from logged data routinely fail to beat the strongest single predictor. Three causes typically go unseparated before more tuning is applied: a mismatched learner, a state that does not predict which model wins, or buffer-to-deployment label shift. A three-stage diagnostic rules them out on a shared buffer. Stage1 estimates a local ceiling on oracle recovery from $k$-NN label consistency. Stage2 asks whether paired BC and offline-RL learners (BC, DQN, and CQL across penalty weights) reach that ceiling. Stage~3 ablates the selector state to test whether richer features would raise it. The combined verdict points to the most promising next step: tuning the learner, redesigning the state, or collecting new data. We apply it to selecting among five dropout-prediction models on edX clickstream data. Across 16 windows, the oracle beats the strongest single base model by 9.7 accuracy points on average, yet BC, DQN, and CQL land in the same test-accuracy band below it (robust to a tenfold buffer sweep and $N{=}2{,}000$ held-out examples). The bottleneck is local representational ambiguity: CQL closes the imitation gap without a deployment gain (not conservatism), regret clusters tightly across learners (not tie-breaking), and the three learners converge on test accuracy (not shift). The next iteration should change the state or collect new data, not tune the offline learner further.

Algorithm Selection (AS) aims to automatically identify the most suitable optimization algorithm for a given problem instance by leveraging measurable problem characteristics and historical performance data. In this study, we investigate the generalization ability of AS models across both synthetic and real-world optimization landscapes. We consider two widely used academic benchmark suites (BBOB and CEC) and two real-world problem sets (robotics trajectory optimization tasks and unmanned aerial vehicle path-planning problems). Through a systematic cross-benchmark evaluation, we analyze how AS models transfer between domains, identify where generalization succeeds or breaks down, and highlight the challenges that arise when applying AS in realistic, domain-specific contexts. Our findings provide insights into the robustness of current AS approaches and inform the development of more reliable, broadly applicable AS systems for real-world optimization.

Small-to-medium scientific datasets place machine learning pipelines under two compounding pressures. Single-run feature selection produces feature sets that change substantially under small perturbations of the training data, and any procedure that uses the same data for selection, tuning, and evaluation produces optimistically biased performance estimates. The two failure modes are routinely treated as separable, but in the regimes where scientific data live, they interact: an unstable selection inflates the variance of an already-optimistic score, and standard remedies for one rarely address the other. RobustModelMaker is a Python framework that couples bootstrap stability selection with strict nested cross-validation, performs all preprocessing and selection inside each fold, and produces a stability-tested feature subset together with a leakage-safe performance estimate. The framework supports nine algorithms across binary classification, multiclass classification, and regression. Behaviour is verified by a deterministic test suite spanning unit, performance, and reproducibility checks on three real scientific datasets comparing to three alternative selectors (ANOVA F-test, recursive feature elimination with cross-validation, and Boruta) on both predictive score and a Jaccard measure of selection stability. RobustModelMaker is competitive in score with the best alternative selector on each dataset, and occupies a position on the joint score-stability frontier that none of the alternatives match across all three task types. Two example applications, ovarian cancer biomarker discovery from the PLCO Trial and critical-temperature regression on the UCI Superconductivity Data, illustrate how the framework is used in practice and what trade-offs become visible when stability is treated as a first-class deliverable rather than an emergent property.

Due to their explicit priors and ability to model uncertainty, Bayesian methods have played a major role in dynamical latent variable modeling of single-cell neural recordings. However, modern-sized datasets have made overparameterized deep networks the preferred methods of choice due to their predictive power and favorable computational scaling. While many posterior approximations exist, all incur approximation errors. Recent work accounts for this error in the form of computational uncertainty but comes at the cost of quadratic complexity and assumes fixed model hyperparameters. Here we extend this development to model selection, including a novel training loss and optimization scheme, which yields tractable inference in large state-spaces. We introduce a framework, the Computation-Aware State-Space Model (CASSM), specifically designed for the scale-imbalanced regime, where the number of trials is significantly lower than the number of recorded neurons. In this regime, for both synthetic and real data, we show that our method is competitive with data-hungry deep networks, with significantly improved uncertainty calibration over previous attempts to scale Bayesian methods. Our experiments provide a roadmap to neuroscience researchers in choosing from a host of potential dynamical latent variable models given key dataset properties and constraints.

We ask: when do Bayesian model averaging (BMA) weights over decision trees carry sufficient epistemic information to justify committed exploitation of the averaging distribution? We answer this question in closed form for Bayesian decision trees (BDTs) with Dirichlet-Multinomial leaf models and a Catalan-exponential tree-size prior (Schetinin&Jakaite, 2025), establishing a complete non-asymptotic theory of rational commitment thresholds.

Automated data science is a structured model-selection problem. A solution must choose data transformations, feature representations, architecture, training procedure, evaluation protocol, and refinement strategy for a task. AutoML systems automate parts of this process, but typically search within predefined pipeline, model, and hyperparameter spaces. LLM-based agents offer greater flexibility through retrieval, code generation, and execution feedback, yet their modelling decisions are often unstructured, difficult to verify, and hard to reuse. We introduce \textsc{MOSAIC} (Modular Orchestration for Structured Agentic Intelligence and Composition), a structured agentic framework for memory-grounded model selection and workflow construction. Given a task and dataset, \textsc{MOSAIC} builds a semantic task profile, retrieves prior cases and source-code modules, and constructs a blueprint: an intermediate representation specifying selected modelling components, composition, interface constraints, and execution requirements. This blueprint turns model selection into a staged, context-grounded search and grounds LLM-based code generation in retrieved evidence rather than unconstrained synthesis. Candidate models are validated by execution and refined using diagnostic feedback, training traces, task metrics, and a failure-aware reinforcement learning policy. We instantiate \textsc{MOSAIC} on financial time-series forecasting and generation, where models must satisfy predictive accuracy, distributional fidelity, execution reliability, and downstream financial criteria such as risk and tail behaviour. Experiments against AutoML and agentic baselines show that \textsc{MOSAIC} improves task performance, execution success, and decision traceability, demonstrating the value of treating automated data science as structured, reusable, and execution-grounded model selection.

Selection bias is a common and often unavoidable aspect of real-world data that challenges the generalizability of machine learning models. When models trained on biased data are deployed in the broader target population, poor model generalization may lead to real harm, particularly in high-risk settings such as healthcare. This risk highlights the need for practitioners to reliably assess model generalizability prior to deployment. However, existing methods for predicting model performance rely on unrealistic access to the target distribution or knowledge of the selection mechanism causing bias. To address these limitations, we propose a novel upper bound on the worst-case model performance on the target population under the realistic setting where the selection mechanism and the target population data are only partially observed. We demonstrate the validity and practical utility of our method through experiments on fully synthetic data, semi-synthetic data derived from the All of Us Research Program, and real-world selection bias in MIMIC-IV. Our work offers a principled and practical tool to estimate the impact of selection bias in an otherwise intractable setting, thereby enabling practitioners to build safer and more generalizable models in healthcare and beyond.

When a learner faces a new task with few samples, it must leverage any available side information. In practice, this often comes in the form of model evaluations on related tasks in public benchmarks. A key question then is how to model task relatedness such that it is both realistic and the benchmark evaluations lead to provable gains. Empirically, we observe that weak monotonicity is often approximately satisfied: if a model dominates another on many benchmarks, it also tends to outperform on the new task. We explore the statistical complexity of learning under (approximate) weak monotonicity, leveraging it within two learning paradigms: transfer learning and model selection aggregation. We show that not only can we prune the model class based on monotonicity, but we can also further adapt to the geometry of the available trade-offs by hedging on the frontier.

Automating scientific computing workflows requires more than generating executable code: autonomous systems must also select appropriate computational strategies, implement them faithfully, and ensure that the resulting outcomes remain causally attributable to the decisions that produced them. In multi-agent pipelines, this process is particularly fragile, as small inconsistencies between agent intentions and actions can lead to semantic drift, where the eventually executed procedure no longer reflects the originally selected strategy, thereby corrupting downstream evaluation and adaptation. In this work, motivated by the ATHENA framework (Toscano et al., 2025; Toscano et al., 2026) and the concept of empowerment (Yiu et al., 2025), we introduce a multi-agent framework that combines contextual bandits with structured inter-agent communication and, most importantly, semantic checkpoints that preserve action-outcome fidelity throughout the pipeline. The system integrates specialized large language model (LLM) agents, grounded code generation, and self-healing execution loops within an adaptive decision-making architecture. Interpreting the framework through the lens of empowerment, we show that reliable autonomous learning requires not only identifying high-quality actions, but also preserving the integrity of their propagation across agents. Using sensitivity analysis and uncertainty quantification workflows as representative case studies, we demonstrate that unchecked semantic drift degrades policy learning, whereas the proposed framework improves convergence, robustness, and adaptation to novel problem contexts. These results suggest a broader design principle for scientific multi-agent systems: adaptive decision-making must be coupled with explicit mechanisms that guarantee semantic consistency and reliable information flow across the computational pipeline.

Given the wide range of deployment targets, flexible model selection is essential for optimizing performance within a given compute budget. Recent work demonstrates that stitching pretrained models within a model family enables cost-effective interpolation of the accuracy-efficiency tradeoff space. Stitching transforms intermediate activations from one pretrained model into another, producing a new interpolated stitched network. Such networks provide a pool of deployment options along the accuracy-efficiency spectrum. However, existing stitching approaches often yield suboptimal tradeoffs and lack generalizability, as they primarily rely on heuristics to select stitch configurations. We argue that constructing improved accuracy-efficiency tradeoffs requires explicitly capturing and leveraging the similarity between pretrained models being stitched. To this end, we introduce KLAS, a novel stitch selection framework that automates and generalizes stitch selection across model families by leveraging KL divergence between intermediate representations. KLAS identifies the most promising binary stitches from the $O(k^2n^2)$ possibilities for $k$ pretrained models of depth $n$. Through comprehensive experiments, we demonstrate that KLAS improves the accuracy-efficiency curve of stitched models at the same finetuning cost as baselines. KLAS achieves up to $1.21\%$ higher ImageNet-1K top-1 accuracy at the same computational cost, or maintains accuracy with a $1.33\times$ reduction in FLOPs.

We study inference in stochastic block models (SBMs) through the lens of optimal transport (OT). We first establish that maximum likelihood variational inference (MLVI) can be interpreted as a semi-relaxed Gromov-Wasserstein (srGW) projection with entropic regularization. While this formulation yields accurate clustering, the entropic regularization prevents transport plans to be sparse, hindering intrinsic model selection. Consequently, we investigate unregularized srGW estimators, and prove that they consistently recover both the SBM connectivity matrix and latent cluster assignments in the asymptotic regime. However, this asymptotic property does not translate into reliable model selection in finite samples, and calls for additional mechanisms to promote sparsity in the inferred cluster proportions. We empirically show that such a regularized formulation yields estimators that simultaneously recover model parameters and select the number of clusters in a single optimization problem, thereby avoiding costly grid search or heuristic model selection procedures.

Maintaining predictive accuracy in non-stationary environments requires online model selection to adapt autonomously to unknown distribution shifts. However, existing tuning-free algorithms face a fundamental trade-off between robustness and agility. Specifically, to ensure dynamic regret bounds, they must restrict learning rates to small constants (e.g., $O(1)$). This restriction inevitably causes significant adaptation lag during abrupt changes. To resolve this, we propose a novel optimistic online mirror descent that utilizes safeguarded large learning rates up to $Θ(T)$, where $T$ is the number of rounds. Our key technical contribution is a post-hoc penalty mechanism that dynamically monitors unstable updates and excludes learning rates incurring excessive regret, eliminating the need for restrictive a priori constraints. We show that the cumulative penalty remains $O(\log T)$, allowing our algorithm to match near-optimal worst-case guarantees while achieving superior rates in benign cases. Empirical evaluations on three synthetic and eleven diverse real-world datasets demonstrate that our approach reduces the adaptation lag from hundreds of rounds to a few rounds, consistently outperforming tuning-free baselines.

This paper addresses structured out-of-distribution (OOD) testing in high-stakes machine learning applications. Traditional conformal methods rely on joint exchangeability, making it difficult to incorporate auxiliary information such as spatiotemporal or grouping structures. To overcome this limitation, we propose the structure-adaptive conformal q-value (SCQ), a significance index that integrates individual test evidence with structural patterns. We also develop pseudo-score-guided transductive automated model selection (P-TAMS), which adapts conformalized model selection to structured OOD testing across a toolbox of candidate models. Together, SCQ and P-TAMS form a unified framework under pairwise exchangeability, providing finite-sample error-rate control, improved power, and enhanced interpretability. Experiments on simulated and real data demonstrate that the proposed approach controls the false discovery rate and performs well across diverse settings.

Recent studies on hallucination detection have shown that hallucination-related signals are more strongly encoded in intermediate layers than in the final layer of large language models (LLMs). Although a growing body of work has sought to exploit this property for hallucination detection, how to automate the selection of high-performing layers remains underexplored, and principled methods for this purpose are still lacking. To address this gap, we first propose several hypotheses for why such signals emerge in intermediate layers and evaluate corresponding criteria for automatic layer selection across diverse LLM architectures, scales, and tasks, covering both question answering and summarization hallucination detection benchmarks. However, we find that none of these criteria consistently delivers satisfactory performance. We therefore propose a new selection criterion, First Effective Peak of Intrinsic Dimension (FEPoID), which consistently identify optimal or near-optimal layers and outperforms both the aforementioned criteria and existing hallucination detection baselines. FEPoID is training-free and incurs negligible computational overhead. In addition, we study the generation behaviors of LLMs and introduce a simple yet effective truncation strategy, which further amplifies hallucination-related signals and substantially improves overall detection performance. Code is publicly available at https://github.com/DesoloYw/Automatic-Layer-Selection-for-Hallucination-Detection.git