Papers

Sequential Model-Based Optimization (SMBO) traditionally relies on Bayesian or ensembling surrogates for uncertainty quantification. While historically treated as fully data-driven, SMBO increasingly integrates external domain expertise to accelerate discovery. To overcome the opaque guidance and diminished integration fidelity of standard acquisition re-weighting, Probabilistic Circuits (PCs) have emerged as a generative surrogate alternative, enabling direct knowledge injection via conditional sampling. However, these generative routines lack the formal exploration-exploitation semantics required for rigorous optimization. We introduce the Distributional Conformance Score (DisCo), a novel metric that unifies the flexibility and efficiency of PCs with a formal uncertainty framework. DisCo provides a bounded, $[0, 1]$-normalized measure of model "surprise" that (1) recovers properties comparable to kernel-based uncertainty known from, e.g., Gaussian Processes, while maintaining linear-time inference, and (2) enables accurate assessment of conformance of external knowledge w.r.t. model evidence. We then present DisCoMBO, a framework leveraging these properties for robust, knowledge-aware optimization. We prove that DisCoMBO is a zero-regret algorithm and demonstrate its effectiveness across diverse benchmarks from AutoML, material optimization, and wind park optimization.

Several tonal pitch spaces and computational models have been proposed to analyze tonal structure in Western tonal music, many of them grounded in principles from music theory and used to support tonal analysis with important implications for tonal tension. In parallel, data-driven methods such as skip-gram have been used to learn chord embeddings from symbolic corpora, but their ability to recover tonal structure and its relation to tonal tension remains underexplored. In this work, we investigate how skip-gram chord embeddings reflect tonal structure and whether they provide a useful basis for analyzing structural aspects of tonal tension. Using chord sequences with and without transposition-based augmentation, we evaluate the learned spaces from geometric, functional, and tension-related perspectives. We show that augmented embeddings exhibit strong transposition equivariance, recover a clear circle-of-fifths structure, and support interpretable shifts between key-related regions of the learned space. We then derive embedding-based measures from chord-to-key distance and contextual chord-distance relations, and show that they capture meaningful aspects of tonal tension structure through correspondence with matched tonal measures and moderate alignment with human tension profiles. Across analyses, transposition-based augmentation generally improves the stability, tonal coherence, and interpretability of the learned space.

Semantic-preserving transformations can induce substantial motion in learned representations, while small changes may strongly affect model predictions, raising a basic question: what local metric best captures semantically consequential variation? We propose Fisher-induced invariant representation geometry (Fisher-IRG), which measures local representation directions through their predictive sensitivity. Around each representation, we construct semantic-preserving and semantic-changing neighborhoods, aggregate their local Fisher information, and recover invariant directions through a contrastive generalized eigenvalue problem. Controlled displacement analyses first show that comparable Euclidean motion can have substantially different predictive consequences, supporting the need for a predictive geometry. Across language and vision models, Fisher-IRG yields stronger semantic-versus-nuisance predictive selectivity and generally more reproducible subspaces than covariance-based geometry, while recovering systematically distinct local directions. Representation interventions further localize semantic effects to the Fisher-derived subspace, and held-out separation and retrieval show that the recovered geometry generalizes beyond the discovery neighborhoods. These results support Fisher-IRG as a principled framework for characterizing local invariant representation geometry.

Large language models are thought to represent features by vectors in a hidden space of dimension given by the model's width. Superposition, in which more features are represented than the width by letting representation vectors overlap, is a leading account of how representation vectors are organized. However, how model width and data statistics determine the configuration of representation vectors and the resulting loss when the number of features and the width are large remains less understood. Here we show, in Anthropic's toy model of superposition, that increasing the width drives a continuous phase transition from a partial-representation phase, where only a subset of features receives appreciable representation vectors while the rest vanish, to a full-representation phase, where every feature is represented. Our theory via a partial random projection approximation predicts, and experiments confirm, that the critical width grows linearly with the number of active features up to a logarithmic factor. The loss scaling changes across the transition: below the critical width, the loss grows linearly with the number of active features and depends weakly on the width in a form set by data statistics; above it, the loss grows approximately quadratically with the number of active features and decays inversely with the width. Non-uniform firing probabilities delay the transition and lower the loss, as more frequent features occupy more space. Our results provide an account of how model width and data statistics jointly shape representations and loss, a step toward understanding representation scaling in large models.

Multivariate time series forecasting (MTSF) is critical across many real-world domains. Existing deep learning approaches fall into two paradigms with distinct limitations: channel-independent (CI) methods unconditionally ignore cross-variable dependencies and model only temporal dynamics, while channel-dependent (CD) methods consider both but typically rely on architectural compromises to mitigate overfitting and computational overhead. We therefore propose Chameleon, a specialized CD state space model (SSM) that enables data-dependent, fine-grained interactions across variables while scaling linearly with their number. By connecting selective SSMs with the Kalman filter, we leverage the missing measurement update in the former for cross-variable modeling while preserving the SSM backbone for robust temporal modeling. We further identify favorable inductive biases of GatedDeltaNet for time series, adapt it as our backbone, and improve generalization through additional techniques, including a previously unexplored stochastic perturbation of reversible instance normalization. On strongly dependent ODE and PEMS datasets, Chameleon achieves the best MSE and MAE across all settings, while its CI ablation and prior CD methods incur 61-178% higher MSE on average. Across 28 standard benchmark settings, Chameleon also achieves better MSE and MAE than each baseline in at least 27 and 22 cases, respectively. Training-time and peak-memory analyses on Traffic and ETT further demonstrate competitive efficiency and favorable memory scalability across different variable counts.

Large language models often preserve meaning despite substantial changes in wording, style, and syntax, while small semantic edits can systematically alter their hidden representations. This suggests that semantic variation may be organized along recurring local directions. We propose the Invariant Atom Hypothesis: local semantic motion admits preferred sparse coordinates along directions that remain stable under meaning-preserving transformations. We learn a shared semantic frame and sparse coordinates that reconstruct semantic displacements while suppressing nuisance variation, with anchor-dependent diagonal modulation adjusting atom strengths without sample-specific rotations. Empirically, the atoms exhibit strong semantic–nuisance separation, sparse reconstruction, reproducible directions, and causal effects on model predictions. The learned geometry generalizes to unseen semantic neighborhoods and nuisance families, while local reweighting improves semantic selectivity and preserves a consistent global-to-local structure. Atom signatures also remain stable under model modification. These findings support reusable invariant directions as a sparse coordinate system for local semantic geometry in language models.

Transformers have demonstrated remarkable in-context learning (ICL) capabilities, enabling them to perform new tasks without additional fine-tuning. However, their performance often deteriorates when encountering out-of-distribution (OOD) inputs that deviate from the training distribution, and the underlying theory remains poorly understood. To fill this gap, we characterize the OOD error under the input distribution shift through the interplay between the dynamics of the so-called $\alpha$-type and $\beta$-type attention weights, which represent the transformer's confidence in identifying the correct and incorrect features, respectively. Our results indicate that the OOD error for each feature depends on all pairwise interactions between the training features and OOD features, and under certain cases the transformer performs no better than random guessing. To improve the OOD generalization performance, we next investigate the impact of model finetuning with the OOD data, and particularly, characterize the model forgetting performance on the source domain. Interestingly, the performance on the source domain may not always degrade after finetuning, which highly depends on the nature of the feature shift: finetuning on OOD domain keeps enhancing the confidence of identifying correct features from the original distribution, while the interference from other incorrect features may either increase or decrease. Extensive experiments on both synthetic and real data are conducted to corroborate the theoretical insights.

A detector pretrained on a broad corpus is fine-tuned on a narrow domain, its in-domain accuracy improves, and it ships. We ask what happens meanwhile to its coverage of objects the vocabulary never names, which in obstacle detection and inspection carry the risk. No in-domain test set holds an example of one. We give a longitudinal protocol: one pretrained checkpoint against its own fine-tuned descendants. It tracks held-out top-$K$ proposal coverage $C_\tau$: of categories pretraining covered and the vocabulary omits, the share of boxes a detector's top $K$ regions still cover. The quantity is the open-world proposal literature's; the longitudinal reading is not. $C_\tau$ falls while in-domain accuracy rises, on four architectures and three domains, by $5.12$ to $63.35$ points on boxes above $1024$ px$^2$. No in-domain number identifies the fall, and neither does detection average precision, which charges a missed and a misnamed box alike. On the one architecture scoring both, adaptation costs $87\%$ of the AP against a fifth of the coverage, and the naming goes first at all six depths of its freeze ladder, every run. What breaks is structured: three architectures sharing no pretraining run agree on which categories lose coverage, and those a model never learned do not lose any. A repair follows and needs no training: mixing a quarter of the pretrained state back, normalisation statistics included, raises coverage on every cell swept for at most $2.47$ points of in-domain accuracy. Seeing it costs one extra evaluation pass.

Reinforcement learning (RL) is a powerful framework for robotic control, yet its practical application is often hindered by high sample complexity. This is particularly restrictive in physical domains where interaction data is costly. While the world often exhibits geometric and physical symmetries, standard RL algorithms typically fail to exploit this structure. In this paper, we demonstrate that exploiting group symmetries significantly reduces the sample complexity of RL. Focusing on finite-horizon Markov decision processes, we find that leveraging homomorphisms induced by group symmetries significantly reduces the theoretical upper and lower bounds on the number of environment interactions required to reach an optimal return. We further extend these bounds to continuous state and action spaces, providing corresponding sample-complexity guarantees under appropriate regularity assumptions. Beyond theory, we validate our findings through controlled experiments and demonstrate the advantages of symmetry-aware policy learning on high-dimensional continuous robotic simulations. Our results show that integrating symmetry into the learning pipeline yields substantial gains in sample efficiency and performance, offering a principled path toward more data-efficient robotics.

Robots operating in real-world environments must often execute complex, multi-step bimanual tasks over long horizons rather than single, isolated actions. Current manipulation datasets provide limited support for this capability: although single-arm datasets reach hundreds of thousands of trajectories, they typically provide only one high-level instruction per episode, while existing bimanual datasets provide subtask annotations for only part of their recorded hours. We present FineART, a densely annotated bimanual manipulation dataset comprising 40,543 episodes (1,718 hours) and 533,913 subtasks across 151 tasks. We also introduce FineART-VLA, a vision-language-action policy that predicts its own next subtask to guide its actions. Mid-training on FineART's subtask annotations improves FineART-VLA's success at following spatial instructions from 32.0% to 100.0%. With step-by-step human subtask guidance, success on unseen long-horizon tasks increases from 16.0% to 76.0%. Furthermore, FineART-VLA matches baseline performance on a new robot with 10x less fine-tuning data and generalizes zero-shot to unseen tasks. We open-source the full dataset, model weights, and training code.

Longitudinal data are valuable because people change. Yet the objectives used to learn from these data can inadvertently erase that change. In person-level contrastive learning, observations from the same person are treated as positives; as records grow, those positives can span increasingly distant—and increasingly different—behavioral states. More history can therefore produce not only more data, but broader invariance. We show that this distinction is fundamental. We separate record span, how much history the learner sees, from supervision span, how far across that history positive-pair supervision reaches. Across in-home sensing records spanning up to 2.7 years, broader supervision systematically suppresses recoverable changing-state information, even when the available history is held fixed. At the broadest span, less than 10% of the information recoverable from an untrained encoder remains. Yet keeping positives local is not sufficient: as records grow, even distant states that are never paired become increasingly similar. Explicitly contrasting other observations from the same person reverses this loss without shortening the record, revealing a second route by which longitudinal scale can broaden invariance. Finally, we prospectively reproduce the supervision-span effect in 199 GLOBEM participants. Longitudinal scale therefore presents a choice: more history need not mean more invariance. By controlling what is held invariant as records grow, we can preserve the change that made the longitudinal data valuable in the first place.

Object detection is a critical application of computer vision, yet models trained uniformly on bounding boxes lack the geometric precision to be fully effective in inference settings, especially at high IoU thresholds. One common approach is the introduction of polygon-supervised localization, where a model is trained explicitly for localization but with access to polygon masks during training. Among such methods, partial polygon supervision is viable due to the higher annotation costs of developing fully supervised training datasets. However, this introduces a further challenge: in object classes of relatively low shape complexity, the addition of training masks and subsequent segmentation loss may obscure box localization, often worsening model performance despite access to deeper geometric information. To address the practical limitations of both full and partial polygon supervision, we introduce a mathematical framework for quantifying per-class geometric complexity by combining information theory and statistical shape analysis. With this framework, we develop Entropy-Aware Geometric Learning Enhancer (EAGLE), a lightweight, post-training model head that learns geometric information for the localization model, improving accuracy via edge refinement and localization rescoring using entropy-informed updates. Empirical analysis across MS COCO and Pascal SBD demonstrates the entropy framework meaningfully predicts shape reconstruction error and accuracy differences among mask-derived and standard box predictions, while the implemented EAGLE framework achieves consistent improvement on already SOTA detection frameworks (+0.9 mAP50-95 points on MS COCO) at minimal observed inference overhead.

Visual navigation requires goal directed planning informed by predictions of environment dynamics. Representative navigation world model methods propose complete action candidates before predicting consequences for selection or search updates. For each candidate, predictions primarily evaluate proposed actions rather than guide its stepwise construction. To this end, we propose InJepa, an intention first planning paradigm centered on latent states. A goal conditioned JEPA proposes latent intentions, and an action conditioned JEPA continually predicts future latent states. Both operate in a shared latent space to align their predictive representations. Within each candidate, predicted consequences feed into the next intention proposal and action inference, forming a closed loop of intention, action, and consequence. This extends the world model's role beyond evaluation to leading candidate construction through latent intentions. We evaluate InJepa through closed loop image goal navigation in Habitat MP3D. The results show effective multistep navigation through intention guided action generation and consequence feedback. Experiments across encoders further support applicability across visual representations. These findings support latent intentions as the central planning object and show that world models can directly guide plan construction through coordinated reasoning about intentions and consequences. Code is available at https://github.com/zhuxun132/InJepa.

World Action Models (WAMs) jointly predict robot actions and future observations. However, their visual prediction objectives do not explicitly distinguish task-relevant transitions from incidental scene changes. For whole-body policies, this distinction matters because task progress depends on coordinated changes in body pose, contact, and object state. We introduce LANGUAGE-FORCING, a training strategy that uses language to emphasize task-relevant changes in the future visual representations of a whole-body WAM. Given paired current and ground- truth short-horizon future frames, a frozen vision-language model assesses task progress and predicts plausible subsequent state changes. We encode these descriptions with a frozen text encoder and align their embeddings with intermediate, action-conditioned future-visual features, complementing the original visual and action prediction objectives. All auxiliary components are used only during training, introducing no additional inference overhead. On HumanoidArena, LANGUAGE-FORCING achieves an average success rate of 79.17%, exceeding the strongest competing policy by 9.46 percentage points and the same WAM without language alignment by 10.77 percentage points. Across five real-world tasks, language alignment raises average success from 53.33% to 71.33% over the same WAM baseline.

Vision-language models are increasingly used for driving-scene understanding, yet their defensive reasoning under weakly structured Global South traffic remains unclear. We introduce BD-HazardVLM, a diagnostic benchmark of 500 Bangladesh road-scene images, comprising 375 from RSUD20K and 125 from TFP-BD, annotated for hazard presence, primary hazard category, risk timing, severity, defensive actions, visible evidence, and hallucination- and Western-road-bias traps. We evaluate Qwen2.5-VL-7B, InternVL3-8B, and MiniCPM-V-4.6 using a shared zero-shot task definition and structured output schema. Binary hazard-presence accuracy ranges from 88.0 percent to 89.2 percent, essentially matching the 89.0 percent majority-class baseline. Fine-grained reasoning is substantially weaker: category accuracy remains below 40 percent, risk-timing accuracy ranges from 1.6 percent to 12.4 percent, and exact defensive-action matching remains below 2 percent. Although InternVL3-8B achieves the highest severity accuracy at 61.2 percent, it has zero recall on the 58 high-severity cases. Qwen exhibits severe category mode collapse, while all models show weak latent-hazard anticipation and near-zero self-reported trap recognition. For the evaluated models, these findings show that coarse hazard recognition does not imply reliable fine-grained defensive-driving reasoning and motivate geographically diverse safety evaluation.

This study investigates generalized boundary representations and transfer learning for physics-informed neural-network (PINN) estimation of spatially varying convective heat transfer coefficients (HTCs) in jet impingement cooling. Piecewise-linear (PWL) and neural-network boundary estimators are inferred from transient interior-temperature measurements and evaluated against CFD-CHT reference HTC profiles at Reynolds numbers of 5000, 10000 and 15000. A discrete Adjoint-FEM formulation with gradient regularization based on the $H^1$-seminorm provides an independent inverse-method comparison. Transfer learning is investigated under two source-target conditions: transfer between noise levels at the same Reynolds number, and transfer from a controlled Gaussian-profile problem to CFD-CHT-derived target problems at different Reynolds numbers. For the representative Re = 5000 target with 5% measurement noise, quadratic, PWL, and neural-network PINNs achieve relative HTC-profile errors of 5.08, 5.31 and 5.19%, respectively, while generalized representations reduce the error for the Re = 15000 target from 16.76% for the quadratic PINN to 14.41% and 14.46%. Transfer learning reduces target-training time by 57.86-81.71% for fixed-Reynolds-number transfer and by 50.59-84.94% for Gaussian-to-CHT transfer. Its effect on reconstruction accuracy is target dependent: fixed-profile PWL transfer remains within 0.30 percentage points of scratch training, whereas Gaussian-to-CHT transfer produces differences of 0.10-1.52 percentage points across targets and representations. These results demonstrate that generalized PINN representations can accommodate spatially varying HTC profiles without a prescribed quadratic form, while transfer learning can substantially reduce target-training cost for repeated inverse estimation, with the resulting reconstruction accuracy depending on source-target compatibility.

A familiar line of reasoning runs as follows: language models are flexible approximators trained under a strictly proper loss, so scale brings them closer to Solomonoff induction, and their predictive success is evidence of that approach. We apply known results from algorithmic information theory, universal prediction and Bayesian consistency. In its uniform form the conclusion holds for no computable model: each has cheap environments it never learns, and every uniformly computable family of total predictors shares one on which each member fails beyond a horizon growing with its index, so “closer” can hold only pointwise, one environment at a time. Nor could the offered routes certify even that: finite evaluation cannot certify agreement on rarely reached contexts, perfect corpus fit selects the corpus law as prior, and asymptotic criteria cannot see the constants that carry Solomonoff's guarantee. Our lens is the Kraft inequality read for predictors. Every model has at most the same budget of predictive success, so Solomonoff's predictor, which leaves part of it unspent, differs from a language model not in having more of it but in where it spends it; on mutually singular families this is exact, and only full-mass Bayes mixtures reach the frontier. We then discuss what is inheritable, including in-context learning as amortised Bayesian inference; what survives within a context window; which Solomonoff-like properties a deployed system should have or avoid; and what served conditionals determine. A closing table states what common practices certify and what to report. Tests are developed in a companion paper.

Since compact, interpretable classifiers correspond to simple programs, it is natural to ask if LLM-based coding agents can be used to learn interpretable models from data. In this paper we present a simple harness that repeatedly prompts a coding LLM to produce the next rule in a rule set, given a sample of labeled examples. We show that this approach is competitive with, and often superior to, state-of-the-art rule-learning methods on tabular data, especially with small training sets, and wins by large margins on text, including against another LLM-based rule learner; that this approach outperforms standard harnesses, and that the approach immediately leads to new kinds of interpretable models, since coding agents are not restricted to search a highly restricted hypothesis space. In addition to the advances made in learning interpretable models, these results are of interest because they introduce a new class of coding benchmarks - construction of interpretable classifiers - which includes quantitative measurements of output quality (i.e., test-set error) and existing mature “specialist coding algorithms” (i.e., existing learning methods).

Masked diffusion language models (MDLMs) decode by repeatedly committing tokens to masked positions, but these commitments are usually irreversible. A token chosen under sparse, partial context is kept fixed, even when later context no longer supports it. Existing samplers mainly decide when to commit a token, but rarely check whether an already committed token should still be kept, allowing early mistakes to propagate. We trace this issue to confidence drift, where the model's confidence in a committed token drops from its sparse commit-time context to the denser context available later. Based on this signal, we propose CoDR (Confidence Drift Remasking), a training-free and sampler-agnostic refinement pass. CoDR estimates drift for all committed positions in only k forward passes via k-partition probing, then remasks and regenerates only the tokens the model no longer endorses. Across two backbones, four reasoning and coding tasks, and three base samplers, CoDR improves average accuracy across all evaluated model-sampler configurations and improves most individual task settings with modest overhead. Controlled experiments show that the gains come from targeted confidence-drift remasking rather than extra compute alone, and that CoDR uses far fewer forward passes than prior remasking methods. Code is available at https://github.com/YueWu0301/CoDR.

A central question in the prediction of stochastic processes is the extent to which past information can improve the probability of correctly predicting the next state. We introduce the Memory Prediction Excess (MPE) to address this question quantitatively. The MPE measures the average improvement in prediction accuracy obtained by using the entire observed history relative to using only the static marginal distribution, in discrete-time finite-state processes. It is defined as the difference between the expected optimal conditional prediction accuracy and the optimal static prediction accuracy. Its basic properties are examined: the MPE is always non-negative; it admits an upper bound depending on the static accuracy, attained if and only if the future is almost surely a deterministic function of the past; and degenerate cases in which the MPE vanishes are characterized. A normalized version, taking values in the unit interval, is introduced as a dimensionless measure of predictive efficiency. A lower bound is derived by comparing predictions based on histories of different lengths, showing that the expected optimal prediction accuracy is monotone with respect to the history length. The framework is extended to finite-length histories, where the finite-history MPE (FH-MPE) measures the predictive gain attainable when only the most recent observations are retained. This leads to the notion of a minimal memory length required to achieve the same predictive performance as the full history. For finite-order Markov chains, this minimal memory length is shown to be bounded by the Markov order. The MPE and its variants are formulated in terms of conditional probabilities and prediction accuracies, offering a probabilistic perspective on the predictive utility of memory that is complementary to classical information-theoretic approaches.

We propose RegLLM, a diagnostic harness for bounded autonomy in regulated agentic workflows. It instruments six trustworthiness signals: citation validity, source grounding, schema compliance, escalation correctness, constitutional alignment, and unsafe-action rate. Signals are distinguished by their source of supervision: programmatic verifiers, task-level escalation labels, or AI-judge scores. A deterministic runtime supervisor blocks ungrounded answers and forces escalation, logging interventions. The same domain constitution informs evaluation, training rewards, and serving guardrails. Task-level should-escalate labels make the act-versus-defer decision a measurable training signal. We demonstrate the harness at smoke scale. An offline reference run (n=12) lifts escalation recall from 0 to 0.67 and reduces unsafe-action rate from 0.33 to 0.08 when governance is enabled. Two single-GPU Qwen2.5-3B LoRA/DPO pilots (n=8, same seed and evaluation split) expose substantial variation: nominally identical RL-base configurations yield task success of 0.25 versus 0.12 and escalation recall of 1.0 versus 0.5. An answer-quality adapter changes recall from 1.0 to 0.5 in Run A, but from 0.5 to 1.0 in Run B. An escalation-aware variant produces no measurable change in Run B. These small pilots do not establish reliable adapter effects or production readiness. Their contribution is diagnostic: configuration variance can overwhelm apparent tuning effects on bounded-autonomy metrics, motivating larger evaluation sets and repeated runs.

On-policy distillation (OPD) trains language models using dense token-level teacher supervision on student-generated trajectories. However, its reliance on frequently refreshed student rollouts often incurs substantial generation cost. We introduce REVO, an off-policy distillation framework that improves rollout efficiency by reusing each student rollout for multi-step learner updates. REVO addresses prefix-level and current-token policy mismatch through stabilized prefix weighting and one-step resampling from the current student, which enables repeated updates without regenerating full trajectories. To prioritize informative token positions within reused rollouts, REVO uses the variance of the student-teacher log-probability ratio to quantify the remaining token-level learning signal and guide repeated optimization. Across multiple student-teacher scales, REVO with only 50 rollout iterations matches or exceeds OPD baselines trained for 200 iterations on both in-domain and cross-domain reasoning benchmarks.

Multiple-choice benchmarks are cheap to grade and are running out of room, and the standard remedy, writing harder items, is slow and repeated for every benchmark. A saturated benchmark still holds a harder task. Each question's wrong options are written for that question alone, so a model can score by eliminating a few options. We propose AnswerPool: take $N$ questions that share a context, pool all their options into one list, and ask the model to assign every question its answer. No item is written and no label changes. The chance of guessing a group right falls from $10^{-3}$ to $5\times10^{-7}$ for five four-option questions, and a model that recognizes its answers keeps its multiple-choice score, so the accuracy lost to pooling measures the credit the format gave for elimination. Deleting answers from the pool makes questions unanswerable with exact ground truth, so abstention is scored in the same pass. Across eight text, image, and video benchmarks and eighteen models, pooling is harder for every model, the elimination credit is largest for the weakest models, and seven of eight open-weight models answer 87 to 100% of unanswerable questions.

Many engineering problems involve optimizing a high-dimensional expensive black-box (HEB) design space. To solve such problems efficiently, we propose a knowledge transfer assisted efficient global optimization (EGO) algorithm, labeled as KT-EGO, which extends the EGO algorithm for solving problems over higher dimensions (i.e., $d>20$). Specifically, the original design space is divided into several low-dimensional subset design spaces. More importantly, in order to extract information from the subset design spaces to accelerate the progress of full optimization, we propose a surrogate-based data fusion strategy in KT-EGO. And further, a searching strategy with an adaptive variable range is devised to enhance the exploitation of promising areas. To show the effectiveness of our proposed algorithm, it is compared against the state-of-the-art algorithms over 12 benchmark functions and a 28-dimensional engineering optimization for the design of compressor blade, which fully validates the effectiveness of the KT-EGO for solving HEB problems.

Deep learning classifiers for dermoscopic skin lesions often reach high in-distribution accuracy while quietly relying on spurious background cues such as skin tone, device vignetting, and embedded rulers, rather than on lesion morphology. This undermines robustness and fairness across skin tones. This work asks whether Explainable AI (XAI), typically used only to audit a finished model, can instead be repurposed as an active training signal that corrects this shortcut without sacrificing diagnostic accuracy. We introduce CAMEO (Class Activation Mapped Equitable Overlay), a framework that improves skin-lesion classification by selecting stable model explanations and using them to separate lesions from their backgrounds. It then replaces the background with realistic synthetic skin while keeping the lesion unchanged. On HAM10000 and dark-skin ISIC images, CAMEO maintained accuracy while reducing background-driven errors by nearly four times. It also made the model's attention more consistent when backgrounds changed. Results across multiple tests show that reducing reliance on background information improves robustness, with Fitzpatrick-based backgrounds providing a realistic and interpretable approach. Results show that XAI-guided augmentation can make dermoscopic classifiers measurably more robust and fair at no cost to accuracy. They also clarify that it is the mechanism and not the specific tone palette that matters, and that the lasting contribution of XAI here lies in stability-screened, annotation-free lesion localisation rather than in the robustness number itself.

As black-box machine learning models become increasingly common, extracting interpretations with uncertainty quantification has become a critical challenge. One popular type of interpretation is leave-one-covariate-out (LOCO) feature importance, while prior LOCO inference methods often require data-splitting or model-refitting. A recent ensemble framework, LOCO-MP, addresses these challenges using minipatches that subsample both observations and features, but massive feature subsampling can hurt prediction in high-dimensional sparse settings. Motivated by this limitation, we consider minipatch ensembles with adaptive feature sampling guided by LOCO importance, and propose LOCO-AdaMP, which enables free LOCO inference for the resulting adaptive minipatch ensemble. We show that LOCO-AdaMP yields substantially improved predictive models while retaining asymptotically valid feature importance inference without data-splitting, despite the complex dependence between the adaptive sampling distribution and the LOCO importance statistics. Our analysis relies on a careful leave-two-out perturbation bound for the iteratively updated sampling probabilities together with the stability of LOCO scores induced by observation subsampling. Empirical results on synthetic and real datasets demonstrate advantages of LOCO-AdaMP over existing methods in predictive performance, inferential power, and stability. Overall, LOCO-AdaMP provides a flexible ensemble framework (agnostic to base models) that delivers both strong predictive performance and asymptotically valid, powerful feature importance inference for regression.

Traditional reinforcement learning (RL) techniques focus on maximizing expected cumulative reward, where each action assumes to take a constant unit of time. However, this assumption does not hold for agentic RL tasks such as machine learning engineering (MLE) agents, where actions involve data loading, feature engineering, and model training that take variable durations. Efficiency matters in modern agentic RL where actions are costly. To address this limitation, we adapt from continuous-time RL and Semi-Markov Decision Process (SMDP) formulation and propose Reward-rate Policy Gradient (RPG), where we focus on optimizing the reward rate – the long-term reward per unit of time. RPG estimates the reward rate from off-policy samples, then charges each action for the time it consumes at that rate. We first conduct theoretical analysis in the bandit setting to establish that RPG approximates the optimal reward rate and empirically demonstrate it outperforms baselines while avoiding enumeration over the policy space, a known issue for an existing method. We then further apply RPG on a small language model (Qwen3.5-4B) with self-improvement loops and empirically show it obtains higher rewards within a fixed time budget than vanilla RL on MLE-Bench and NanoGPT, with a 19.2% and 85.7% margin, respectively. Our method provides a practical solution for optimizing performance under wait time considerations in modern agentic RL tasks, where actions interact with external environments and cost time.

An ongoing challenge in sensory neuroscience is to characterize the feature dimensions encoded by cortical populations. Recent approaches probe feature selectivity in a data-driven way, by synthesizing a most-exciting-input (MEI) for a target neural population. While this approach has been successfully applied to human higher visual cortex using fMRI data, generating MEIs for early- and mid-level retinotopic visual areas requires additional modeling constraints due to small receptive field sizes. To address this challenge, we introduce two novel MEI generation frameworks, Receptive Field Diffusion for Visual Exploration (RF-DiVE) and Receptive Field Gradient Optimization (RF-GO). Both methods use a population receptive field (pRF)-constrained voxelwise encoding model; RF-DiVE combines this with a pretrained latent diffusion model, while RF-GO uses regularized gradient ascent. When applied to single voxels in retinotopically defined areas V1-hV4, using data from the Natural Scenes Dataset, we obtain MEIs that exhibit consistent structure within the pRF, suggesting selectivity for local features like contour, color, and texture. We systematically compare MEIs generated by RF-DiVE and RF-GO using two encoding backbones, performing in-silico validation of predicted responses to MEIs using independent encoding models. Across all methods and all visual areas, MEIs elicit higher model-predicted responses than the most activating natural images. We further find that the choice of generation framework and encoding backbone differentially affects MEI properties, including their visual appearance, structural interpretability, and cross-model generalizability. These results offer a new approach for performing data-driven characterization of spatial and feature selectivity across human visual cortex.

Offline reinforcement learning seeks optimal decision rules from previously collected data. In some applications, a decision can be an entire function, such as a fluence map in radiation therapy or a smooth movement trajectory in robotics. In this paper, we study the finite-sample theory for fitted Q-iteration (FQI) with functional actions in a discounted infinite-horizon setting. Three major difficulties arise in this setting: first, the absence of a Lebesgue probability density for functional actions complicates coverage descriptions; second, conventional coverage requirements can be restrictive; and third, the large functional action space makes greedy optimization in FQI challenging. To address these difficulties, we study smoothness-regularized policy search under a critic-relative coverage condition. This condition measures how well logged data distinguish relevant action-value differences without requiring an action density. Our main theorem gives finite-sample guarantees for learned-policy regret relative to the best value within a fixed smooth class of functional-action policies. The results allow trajectory lengths to be either bounded or growing and Q-functions to be fitted by either functional-input kernel ridge regression or adaptive functional neural networks. For a few examples, we can obtain polynomially decaying regret bounds in the number of logged transitions, up to logarithmic factors, with logarithmically many FQI iterations. Numerical experiments show gains of learned functional-action policies over constant-action policies and support our adoption of a critic-relative coverage condition.

Continual learning is usually studied through mechanisms that preserve old knowledge. We develop Successional Learning Theory (SLT), a mesoscopic account in which the current representation is a resident community, the incoming task is an invader, forgetting is resident displacement, joint retention is coexistence, replay is resident reinforcement, and training moves from establishment toward stabilization. Its empirical coordinate is directional pre-invasion compatibility, measured on the resident model before the incoming task is learned. Across eight experiments, compatibility orders later forgetting on the 20 directed Split-CIFAR-10 transitions (three-repeat r=-0.789, incoming-task cluster 95% CI [-0.90,-0.72], every repeat alone r<=-0.67), forecasts held-out forgetting with 24% lower error than a no-information baseline, and reproduces under controlled MNIST permutations and CIFAR-10 rotations (r=-0.804, -0.718). On an 84-transition suite, compatibility separates coexistence from exclusion at every retention threshold (AUC 0.93-0.97). Replay repairs every transition with at most 325 stored examples and is most efficient where displacement is largest. Compatibility reaches |r|=0.720, while activation, representation, Jacobian, and fixed-coefficient Lotka-Volterra specializations do not. Plasticity and feature turnover fall reliably from early to late training (15/15 and 14/15 runs). We formalize a minimum habitat-modification bound, a displacement floor, a sufficient coexistence condition, an identifiability law with a range-restriction corollary, successional stabilization, and local reinforcement. The identifiability law also predicts where the coordinate loses leverage, and the prediction matches three CIFAR-100 partitions and five optimizer regimes. SLT is a pre-adaptation diagnostic that complements replay, regularization, and projection methods.