Human cognition is deeply intertwined with a sense of time, known as Chronoception. This sense allows us to judge how long facts remain valid and when knowledge becomes outdated. Despite progress in vision, language, and motor control, AI still struggles to reason about temporal validity. We introduce Chronocept, the first benchmark to model temporal validity as a continuous probability distribution over time. Using skew-normal curves fitted along semantically decomposed temporal axes, Chronocept captures nuanced patterns of emergence, decay, and peak relevance. It includes two datasets: Benchmark I (atomic facts) and Benchmark II (multi-sentence passages). Annotations show strong inter-annotator agreement (84% and 89%). Our baselines predict curve parameters - location, scale, and skewness - enabling interpretable, generalizable learning and outperforming classification-based approaches. Chronocept fills a foundational gap in AI's temporal reasoning, supporting applications in knowledge grounding, fact-checking, retrieval-augmented generation (RAG), and proactive agents. Code and data are publicly available.
The primary goal of this study is to develop and evaluate an innovative prompting technique, AnaQuest, for generating multiple-choice questions (MCQs) using a pre-trained large language model. In AnaQuest, the choice items are sentence-level assertions about complex concepts. The technique integrates formative and summative assessments. In the formative phase, students answer open-ended questions for target concepts in free text. For summative assessment, AnaQuest analyzes these responses to generate both correct and incorrect assertions. To evaluate the validity of the generated MCQs, Item Response Theory (IRT) was applied to compare item characteristics between MCQs generated by AnaQuest, a baseline ChatGPT prompt, and human-crafted items. An empirical study found that expert instructors rated MCQs generated by both AI models to be as valid as those created by human instructors. However, IRT-based analysis revealed that AnaQuest-generated questions - particularly those with incorrect assertions (foils) - more closely resembled human-crafted items in terms of difficulty and discrimination than those produced by ChatGPT.
Large Language Models (LLMs) have unveiled remarkable capabilities in understanding and generating both natural language and code, but LLM reasoning is prone to hallucination and struggle with complex, novel scenarios, often getting stuck on partial or incorrect solutions. However, the inherent ability of Evolutionary Algorithms (EAs) to explore extensive and complex search spaces makes them particularly effective in scenarios where traditional optimization methodologies may falter. However, EAs explore a vast search space when applied to complex problems. To address the computational bottleneck of evaluating large populations, particularly crucial for complex evolutionary tasks, we introduce a highly efficient evaluation framework. This implementation maintains compatibility with existing primitive definitions, ensuring the generation of valid individuals. Using LLMs, we propose an enhanced evolutionary search strategy that enables a more focused exploration of expansive solution spaces. LLMs facilitate the generation of superior candidate solutions, as evidenced by empirical results demonstrating their efficacy in producing improved outcomes.
Scientific diagrams are vital tools for communicating structured knowledge across disciplines. However, they are often published as static raster images, losing symbolic semantics and limiting reuse. While Multimodal Large Language Models (MLLMs) offer a pathway to bridging vision and structure, existing methods lack semantic control and structural interpretability, especially on complex diagrams. We propose Draw with Thought (DwT), a training-free framework that guides MLLMs to reconstruct diagrams into editable mxGraph XML code through cognitively-grounded Chain-of-Thought reasoning. DwT enables interpretable and controllable outputs without model fine-tuning by dividing the task into two stages: Coarse-to-Fine Planning, which handles perceptual structuring and semantic specification, and Structure-Aware Code Generation, enhanced by format-guided refinement. To support evaluation, we release Plot2XML, a benchmark of 247 real-world scientific diagrams with gold-standard XML annotations. Extensive experiments across eight MLLMs show that our approach yields high-fidelity, semantically aligned, and structurally valid reconstructions, with human evaluations confirming strong alignment in both accuracy and visual aesthetics, offering a scalable solution for converting static visuals into executable representations and advancing machine understanding of scientific graphics.
With the rise of AI-based code generation, customizing existing code out of natural language instructions to modify visual results -such as figures or images -has become possible, promising to reduce the need for deep programming expertise. However, even experienced developers can struggle with this task, as it requires identifying relevant code regions (feature location), generating valid code variants, and ensuring the modifications reliably align with user intent. In this paper, we introduce vTikZ, the first benchmark designed to evaluate the ability of Large Language Models (LLMs) to customize code while preserving coherent visual outcomes. Our benchmark consists of carefully curated vTikZ editing scenarios, parameterized ground truths, and a reviewing tool that leverages visual feedback to assess correctness. Empirical evaluation with stateof-the-art LLMs shows that existing solutions struggle to reliably modify code in alignment with visual intent, highlighting a gap in current AI-assisted code editing approaches. We argue that vTikZ opens new research directions for integrating LLMs with visual feedback mechanisms to improve code customization tasks in various domains beyond TikZ, including image processing, art creation, Web design, and 3D modeling.
We introduce a framework for robust uncertainty quantification in situations where labeled training data are corrupted, through noisy or missing labels. We build on conformal prediction, a statistical tool for generating prediction sets that cover the test label with a pre-specified probability. The validity of conformal prediction, however, holds under the i.i.d assumption, which does not hold in our setting due to the corruptions in the data. To account for this distribution shift, the privileged conformal prediction (PCP) method proposed leveraging privileged information (PI) -- additional features available only during training -- to re-weight the data distribution, yielding valid prediction sets under the assumption that the weights are accurate. In this work, we analyze the robustness of PCP to inaccuracies in the weights. Our analysis indicates that PCP can still yield valid uncertainty estimates even when the weights are poorly estimated. Furthermore, we introduce uncertain imputation (UI), a new conformal method that does not rely on weight estimation. Instead, we impute corrupted labels in a way that preserves their uncertainty. Our approach is supported by theoretical guarantees and validated empirically on both synthetic and real benchmarks. Finally, we show that these techniques can be integrated into a triply robust framework, ensuring statistically valid predictions as long as at least one underlying method is valid.
Trustworthy depression prediction based on deep learning, incorporating both predictive reliability and algorithmic fairness across diverse demographic groups, is crucial for clinical application. Recently, achieving reliable depression predictions through uncertainty quantification has attracted increasing attention. However, few studies have focused on the fairness of uncertainty quantification (UQ) in depression prediction. In this work, we investigate the algorithmic fairness of UQ, namely Equal Opportunity Coverage (EOC) fairness, and propose Fair Uncertainty Quantification (FUQ) for depression prediction. FUQ pursues reliable and fair depression predictions through group-based analysis. Specifically, we first group all the participants by different sensitive attributes and leverage conformal prediction to quantify uncertainty within each demographic group, which provides a theoretically guaranteed and valid way to quantify uncertainty for depression prediction and facilitates the investigation of fairness across different demographic groups. Furthermore, we propose a fairness-aware optimization strategy that formulates fairness as a constrained optimization problem under EOC constraints. This enables the model to preserve predictive reliability while adapting to the heterogeneous uncertainty levels across demographic groups, thereby achieving optimal fairness. Through extensive evaluations on several visual and audio depression datasets, our approach demonstrates its effectiveness.
We introduce the novel task of Language-Guided Object Placement in Real 3D Scenes. Our model is given a 3D scene's point cloud, a 3D asset, and a textual prompt broadly describing where the 3D asset should be placed. The task here is to find a valid placement for the 3D asset that respects the prompt. Compared with other language-guided localization tasks in 3D scenes such as grounding, this task has specific challenges: it is ambiguous because it has multiple valid solutions, and it requires reasoning about 3D geometric relationships and free space. We inaugurate this task by proposing a new benchmark and evaluation protocol. We also introduce a new dataset for training 3D LLMs on this task, as well as the first method to serve as a non-trivial baseline. We believe that this challenging task and our new benchmark could become part of the suite of benchmarks used to evaluate and compare generalist 3D LLM models.
Boundary representation (B-rep) of geometric models is a fundamental format in Computer-Aided Design (CAD). However, automatically generating valid and high-quality B-rep models remains challenging due to the complex interdependence between the topology and geometry of the models. Existing methods tend to prioritize geometric representation while giving insufficient attention to topological constraints, making it difficult to maintain structural validity and geometric accuracy. In this paper, we propose DTGBrepGen, a novel topology-geometry decoupled framework for B-rep generation that explicitly addresses both aspects. Our approach first generates valid topological structures through a two-stage process that independently models edge-face and edge-vertex adjacency relationships. Subsequently, we employ Transformer-based diffusion models for sequential geometry generation, progressively generating vertex coordinates, followed by edge geometries and face geometries which are represented as B-splines. Extensive experiments on diverse CAD datasets show that DTGBrepGen significantly outperforms existing methods in both topological validity and geometric accuracy, achieving higher validity rates and producing more diverse and realistic B-reps. Our code is publicly available at https://github.com/jinli99/DTGBrepGen.
When solving optimization problems under uncertainty with contextual data, utilizing machine learning to predict the uncertain parameters is a popular and effective approach. Decision-focused learning (DFL) aims at learning a predictive model such that decision quality, instead of prediction accuracy, is maximized. Common practice here is to predict a single value for each uncertain parameter, implicitly assuming that there exists a (single-scenario) deterministic problem approximation (proxy) that is sufficient to obtain an optimal decision. Other work assumes the opposite, where the underlying distribution needs to be estimated. However, little is known about when either choice is valid. This paper investigates for the first time problem properties that justify using either assumption. Using this, we present effective decision proxies for DFL, with very limited compromise on the complexity of the learning task. We show the effectiveness of presented approaches in experiments on problems with continuous and discrete variables, as well as uncertainty in the objective function and in the constraints.
The ability to selectively attend to relevant stimuli while filtering out distractions is essential for agents that process complex, high-dimensional sensory input. This paper introduces a model of covert and overt visual attention through the framework of active inference, utilizing dynamic optimization of sensory precisions to minimize free-energy. The model determines visual sensory precisions based on both current environmental beliefs and sensory input, influencing attentional allocation in both covert and overt modalities. To test the effectiveness of the model, we analyze its behavior in the Posner cueing task and a simple target focus task using two-dimensional(2D) visual data. Reaction times are measured to investigate the interplay between exogenous and endogenous attention, as well as valid and invalid cueing. The results show that exogenous and valid cues generally lead to faster reaction times compared to endogenous and invalid cues. Furthermore, the model exhibits behavior similar to inhibition of return, where previously attended locations become suppressed after a specific cue-target onset asynchrony interval. Lastly, we investigate different aspects of overt attention and show that involuntary, reflexive saccades occur faster than intentional ones, but at the expense of adaptability.
Many inverse problems in nuclear fusion and high-energy astrophysics research, such as the optimization of tokamak reactor geometries or the inference of black hole parameters from interferometric images, necessitate high-dimensional parameter scans and large ensembles of simulations to be performed. Such inverse problems typically involve large uncertainties, both in the measurement parameters being inverted and in the underlying physics models themselves. Monte Carlo sampling, when combined with modern non-linear dimensionality reduction techniques such as autoencoders and manifold learning, can be used to reduce the size of the parameter spaces considerably. However, there is no guarantee that the resulting combinations of parameters will be physically valid, or even mathematically consistent. In this position paper, we advocate adopting a hybrid approach that leverages our recent advances in the development of formal verification methods for numerical algorithms, with the goal of constructing parameter space restrictions with provable mathematical and physical correctness properties, whilst nevertheless respecting both experimental uncertainties and uncertainties in the underlying physical processes.
Estimating treatment effects from observational data is challenging due to two main reasons: (a) hidden confounding, and (b) covariate mismatch (control and treatment groups not having identical distributions). Long lines of works exist that address only either of these issues. To address the former, conventional techniques that require detailed knowledge in the form of causal graphs have been proposed. For the latter, covariate matching and importance weighting methods have been used. Recently, there has been progress in combining testable independencies with partial side information for tackling hidden confounding. A common framework to address both hidden confounding and selection bias is missing. We propose neural architectures that aim to learn a representation of pre-treatment covariates that is a valid adjustment and also satisfies covariate matching constraints. We combine two different neural architectures: one based on gradient matching across domains created by subsampling a suitable anchor variable that assumes causal side information, followed by the other, a covariate matching transformation. We prove that approximately invariant representations yield approximate valid adjustment sets which would enable an interval around the true causal effect. In contrast to usual sensitivity analysis, where an unknown nuisance parameter is varied, we have a testable approximation yielding a bound on the effect estimate. We also outperform various baselines with respect to ATE and PEHE errors on causal benchmarks that include IHDP, Jobs, Cattaneo, and an image-based Crowd Management dataset.
We live in a multivariate world, and effective modeling of financial portfolios, including their construction, allocation, forecasting, and risk analysis, simply is not possible without explicitly modeling the dependence structure of their assets. Dependence structure can drive portfolio results more than the combined effects of other parameters in investment and risk models, but the literature provides relatively little to define the finite-sample distributions of dependence measures in useable and useful ways under challenging, real-world financial data conditions. Yet this is exactly what is needed to make valid inferences about their estimates, and to use these inferences for essential purposes such as hypothesis testing, dynamic monitoring, realistic and granular scenario and reverse scenario analyses, and mitigating the effects of correlation breakdowns during market upheavals. This work develops a new and straightforward method, Nonparametric Angles-based Correlation (NAbC), for defining the finite-sample distributions of any dependence measure whose matrix of pairwise associations is positive definite (e.g. Pearsons, Kendalls, Spearmans, Tail Dependence Matrix, and others). The solution remains valid under marginal asset distributions characterized by notably different and varying degrees of serial correlation, non-stationarity, heavy-tailedness, and asymmetry. Importantly, NAbCs p-values and confidence intervals remain analytically consistent at both the matrix level and the pairwise cell level. Finally, NAbC maintains validity even when selected cells in the matrix are frozen for a given scenario or stress test, thus enabling flexible, granular, and realistic scenarios. NAbC stands alone in providing all of these capabilities simultaneously, and should prove to be a very useful means by which we can better understand and manage financial portfolios in our multivariate world.
Integrated sensing and communication (ISAC) poses various challenges that arise from the communication-centric design of cellular networks. One of them is target detection with time division duplex (TDD) transmission used in current 5G and future 6G deployments, where the periodic on-off behavior of the transmitter creates impulsive sidelobes in the radar point spread function (PSF). These can be mistaken for actual targets by conventional peak detection techniques, leading to false alarms. In this work, we first analytically describe the range-Doppler PSF due to TDD windowing. We then propose a computationally efficient method that leverages the PSF to distinguish impulsive sidelobes from valid target peaks. Simulation results and outdoor drone measurements with an ISAC proof of concept demonstrate the capability of our algorithm, showing that it can achieve reliable target detection while limiting false alarms.
We propose new reproducing kernel-based tests for model checking in conditional moment restriction models. By regressing estimated residuals on kernel functions via kernel ridge regression (KRR), we obtain a coefficient function in a reproducing kernel Hilbert space (RKHS) that is zero if and only if the model is correctly specified. We introduce two classes of test statistics: (i) projection-based tests, using RKHS inner products to capture global deviations, and (ii) random location tests, evaluating the KRR estimator at randomly chosen covariate points to detect local departures. The tests are consistent against fixed alternatives and sensitive to local alternatives at the $n^{-1/2}$ rate. When nuisance parameters are estimated, Neyman orthogonality projections ensure valid inference without repeated estimation in bootstrap samples. The random location tests are interpretable and can visualize model misspecification. Simulations show strong power and size control, especially in higher dimensions, outperforming existing methods.
This paper proposes an optimal policy that targets the average welfare of the worst-off $\alpha$-fraction of the post-treatment outcome distribution. We refer to this policy as the $\alpha$-Expected Welfare Maximization ($\alpha$-EWM) rule, where $\alpha \in (0,1]$ denotes the size of the subpopulation of interest. The $\alpha$-EWM rule interpolates between the expected welfare ($\alpha=1$) and the Rawlsian welfare ($\alpha\rightarrow 0$). For $\alpha\in (0,1)$, an $\alpha$-EWM rule can be interpreted as a distributionally robust EWM rule that allows the target population to have a different distribution than the study population. Using the dual formulation of our $\alpha$-expected welfare function, we propose a debiased estimator for the optimal policy and establish its asymptotic upper regret bounds. In addition, we develop asymptotically valid inference for the optimal welfare based on the proposed debiased estimator. We examine the finite sample performance of the debiased estimator and inference via both real and synthetic data.
Counterfactual mean estimators such as difference-in-differences and synthetic control have grown into workhorse tools for program evaluation. Inference for these estimators is well-developed in settings where all post-treatment data is available at the time of analysis. However, in settings where data arrives sequentially, these tests do not permit real-time inference, as they require a pre-specified sample size T. We introduce real-time inference for program evaluation through anytime-valid rank tests. Our methodology relies on interpreting the absence of a treatment effect as exchangeability of the treatment estimates. We then convert these treatment estimates into sequential ranks, and construct optimal finite-sample valid sequential tests for exchangeability. We illustrate our methods in the context of difference-in-differences and synthetic control. In simulations, they control size even under mild exchangeability violations. While our methods suffer slight power loss at T, they allow for early rejection (before T) and preserve the ability to reject later (after T).
This paper analyzes a model in which an outcome variable equals a frontier function of inputs less a nonnegative unobserved deviation. If zero lies in the support of the deviation given inputs, then the frontier function is identified by the supremum outcome at those inputs. This obviates the need for instrumental variables. Implementation requires allowing for the distribution of deviations to depend on inputs, thus not ruling out endogeneity. Including random errors yields a stochastic frontier analysis model. We generalize this model to allow the joint distribution of deviations and errors to depend on inputs. We derive a lower bound for the mean deviation based only on variance and skewness, requiring no other distributional assumptions and remaining valid even when zero is not in the support of deviations or data are sparse near the frontier. We apply our results to a frontier production function, where deviations represent inefficiencies.
Transfer learning (TL) for high-dimensional regression (HDR) is an important problem in machine learning, particularly when dealing with limited sample size in the target task. However, there currently lacks a method to quantify the statistical significance of the relationship between features and the response in TL-HDR settings. In this paper, we introduce a novel statistical inference framework for assessing the reliability of feature selection in TL-HDR, called PTL-SI (Post-TL Statistical Inference). The core contribution of PTL-SI is its ability to provide valid $p$-values to features selected in TL-HDR, thereby rigorously controlling the false positive rate (FPR) at desired significance level $\alpha$ (e.g., 0.05). Furthermore, we enhance statistical power by incorporating a strategic divide-and-conquer approach into our framework. We demonstrate the validity and effectiveness of the proposed PTL-SI through extensive experiments on both synthetic and real-world high-dimensional datasets, confirming its theoretical properties and utility in testing the reliability of feature selection in TL scenarios.