We propose a machine-learning tool that yields causal inference on text in randomized trials. Based on a simple econometric framework in which text may capture outcomes of interest, our procedure addresses three questions: First, is the text affected by the treatment? Second, which outcomes is the effect on? And third, how complete is our description of causal effects? To answer all three questions, our approach uses large language models (LLMs) that suggest systematic differences across two groups of text documents and then provides valid inference based on costly validation. Specifically, we highlight the need for sample splitting to allow for statistical validation of LLM outputs, as well as the need for human labeling to validate substantive claims about how documents differ across groups. We illustrate the tool in a proof-of-concept application using abstracts of academic manuscripts.
Uncertainty quantification is necessary for developers, physicians, and regulatory agencies to build trust in machine learning predictors and improve patient care. Beyond measuring uncertainty, it is crucial to express it in clinically meaningful terms that provide actionable insights. This work introduces a conformal risk control (CRC) procedure for organ-dependent uncertainty estimation, ensuring high-probability coverage of the ground-truth image. We first present a high-dimensional CRC procedure that leverages recent ideas of length minimization. We make this procedure semantically adaptive to each patient's anatomy and positioning of organs. Our method, sem-CRC, provides tighter uncertainty intervals with valid coverage on real-world computed tomography (CT) data while communicating uncertainty with clinically relevant features.
Fuzz testing is a crucial component of software security assessment, yet its effectiveness heavily relies on valid fuzz drivers and diverse seed inputs. Recent advancements in Large Language Models (LLMs) offer transformative potential for automating fuzz testing (LLM4Fuzz), particularly in generating drivers and seeds. However, current LLM4Fuzz solutions face critical reliability challenges, including low driver validity rates and seed quality trade-offs, hindering their practical adoption. This paper aims to examine the reliability bottlenecks of LLM-driven fuzzing and explores potential research directions to address these limitations. It begins with an overview of the current development of LLM4SE and emphasizes the necessity for developing reliable LLM4Fuzz solutions. Following this, the paper envisions a vision where reliable LLM4Fuzz transforms the landscape of software testing and security for industry, software development practitioners, and economic accessibility. It then outlines a road ahead for future research, identifying key challenges and offering specific suggestions for the researchers to consider. This work strives to spark innovation in the field, positioning reliable LLM4Fuzz as a fundamental component of modern software testing.
Logical reasoning is essential for large language models (LLMs) to ensure accurate and coherent inference. However, LLMs struggle with reasoning order variations and fail to generalize across logically equivalent transformations. LLMs often rely on fixed sequential patterns rather than true logical understanding. To address this issue, we introduce an order-centric data augmentation framework based on commutativity in logical reasoning. We first randomly shuffle independent premises to introduce condition order augmentation. For reasoning steps, we construct a directed acyclic graph (DAG) to model dependencies between steps, which allows us to identify valid reorderings of steps while preserving logical correctness. By leveraging order-centric augmentations, models can develop a more flexible and generalized reasoning process. Finally, we conduct extensive experiments across multiple logical reasoning benchmarks, demonstrating that our method significantly enhances LLMs' reasoning performance and adaptability to diverse logical structures. We release our codes and augmented data in https://anonymous.4open.science/r/Order-Centric-Data-Augmentation-822C/.
Conventional survival metrics, such as Harrell's concordance index (CI) and the Brier Score, rely on the independent censoring assumption for valid inference with right-censored data. However, in the presence of so-called dependent censoring, where the probability of censoring is related to the event of interest, these metrics can give biased estimates of the underlying model error. In this paper, we introduce three new evaluation metrics for survival analysis based on Archimedean copulas that can account for dependent censoring. We also develop a framework to generate realistic, semi-synthetic datasets with dependent censoring to facilitate the evaluation of the metrics. Our experiments in synthetic and semi-synthetic data demonstrate that the proposed metrics can provide more accurate estimates of the model error than conventional metrics under dependent censoring.
Conformal predictors provide set or functional predictions that are valid under the assumption of randomness, i.e., under the assumption of independent and identically distributed data. The question asked in this paper is whether there are predictors that are valid in the same sense under the assumption of randomness and that are more efficient than conformal predictors. The answer is that the class of conformal predictors is universal in that only limited gains in predictive efficiency are possible. The previous work in this area has relied on the algorithmic theory of randomness and so involved unspecified constants, whereas this paper's results are much more practical. They are also shown to be optimal in some respects.
Traditional psychological experiments utilizing naturalistic stimuli face challenges in manual annotation and ecological validity. To address this, we introduce a novel paradigm leveraging multimodal large language models (LLMs) as proxies to extract rich semantic information from naturalistic images through a Visual Question Answering (VQA) strategy for analyzing human visual semantic representation. LLM-derived representations successfully predict established neural activity patterns measured by fMRI (e.g., faces, buildings), validating its feasibility and revealing hierarchical semantic organization across cortical regions. A brain semantic network constructed from LLM-derived representations identifies meaningful clusters reflecting functional and contextual associations. This innovative methodology offers a powerful solution for investigating brain semantic organization with naturalistic stimuli, overcoming limitations of traditional annotation methods and paving the way for more ecologically valid explorations of human cognition.
Large language models (LLMs) are often deployed to perform constrained tasks, with narrow domains. For example, customer support bots can be built on top of LLMs, relying on their broad language understanding and capabilities to enhance performance. However, these LLMs are adversarially susceptible, potentially generating outputs outside the intended domain. To formalize, assess, and mitigate this risk, we introduce domain certification; a guarantee that accurately characterizes the out-of-domain behavior of language models. We then propose a simple yet effective approach, which we call VALID that provides adversarial bounds as a certificate. Finally, we evaluate our method across a diverse set of datasets, demonstrating that it yields meaningful certificates, which bound the probability of out-of-domain samples tightly with minimum penalty to refusal behavior.
Generalization to novel compound tasks under distribution shift is important for deploying transformer-based language models (LMs). This work investigates Chain-of-Thought (CoT) reasoning as a means to enhance OOD generalization. Through controlled experiments across several compound tasks, we reveal three key insights: (1) While QA-trained models achieve near-perfect in-distribution accuracy, their OOD performance degrades catastrophically, even with 10000k+ training examples; (2) the granularity of CoT data strongly correlates with generalization performance; finer-grained CoT data leads to better generalization; (3) CoT exhibits remarkable sample efficiency, matching QA performance with much less (even 80%) data. Theoretically, we demonstrate that compound tasks inherently permit shortcuts in Q-A data that misalign with true reasoning principles, while CoT forces internalization of valid dependency structures, and thus can achieve better generalization. Further, we show that transformer positional embeddings can amplify generalization by emphasizing subtask condition recurrence in long CoT sequences. Our combined theoretical and empirical analysis provides compelling evidence for CoT reasoning as a crucial training paradigm for enabling LM generalization under real-world distributional shifts for compound tasks.
We investigate the use of large language models (LLMs) to simulate human responses to survey questions, and perform uncertainty quantification to gain reliable insights. Our approach converts imperfect LLM-simulated responses into confidence sets for population parameters of human responses, addressing the distribution shift between the simulated and real populations. A key innovation lies in determining the optimal number of simulated responses: too many produce overly narrow confidence sets with poor coverage, while too few yield excessively loose estimates. To resolve this, our method adaptively selects the simulation sample size, ensuring valid average-case coverage guarantees. It is broadly applicable to any LLM, irrespective of its fidelity, and any procedure for constructing confidence sets. Additionally, the selected sample size quantifies the degree of misalignment between the LLM and the target human population. We illustrate our method on real datasets and LLMs.
Control Barrier Functions (CBFs) offer a framework for ensuring set invariance and designing constrained control laws. However, crafting a valid CBF relies on system-specific assumptions and the availability of an accurate system model, underscoring the need for systematic data-driven synthesis methods. This paper introduces a data-driven approach to synthesizing a CBF for discrete-time LTI systems using only input-output measurements. The method begins by computing the maximal control invariant set using an input-output data-driven representation, eliminating the need for precise knowledge of the system's order and explicit state estimation. The proposed CBF is then systematically derived from this set, which can accommodate multiple input-output constraints. Furthermore, the proposed CBF is leveraged to develop a minimally invasive safety filter that ensures recursive feasibility with an adaptive decay rate. To improve clarity, we assume a noise-free dataset, though the method can be extended using data-driven reachability to capture noise effects and handle uncertainty. Finally, the effectiveness of the proposed method is demonstrated on an unknown time-delay system.
We use valid inequalities (cuts) of the binary integer program for winner determination in a combinatorial auction (CA) as "artificial items" that can be interpreted intuitively and priced to generate Artificial Walrasian Equilibria. While the lack of an integer programming gap is sufficient to guarantee a Walrasian equilibrium, we show that it does not guarantee a "price-match equilibrium" (PME), a refinement that we introduce, in which prices are justified by an iso-revenue outcome for any hypothetical removal of a single bidder. We prove the existence of PME for any CA and characterize their economic properties and computation. We implement minimally artificial PME rules and compare them with other prominent CA payment rules in the literature.
Large language models (LLMs) have demonstrated impressive performance in various natural language processing tasks, yet their ability to perform multi-step logical reasoning remains an open challenge. Although Chain-of-Thought prompting has improved logical reasoning by enabling models to generate intermediate steps, it lacks mechanisms to assess the coherence of these logical transitions. In this paper, we propose a novel, lightweight evaluation strategy for logical reasoning that uses query-key alignments inside transformer attention heads. By computing a single forward pass and extracting a "QK-score" from carefully chosen heads, our method reveals latent representations that reliably separate valid from invalid inferences, offering a scalable alternative to traditional ablation-based techniques. We also provide an empirical validation on multiple logical reasoning benchmarks, demonstrating improved robustness of our evaluation method against distractors and increased reasoning depth. The experiments were conducted on a diverse set of models, ranging from 1.5B to 70B parameters.
Event argument extraction identifies arguments for predefined event roles in text. Traditional evaluations rely on exact match (EM), requiring predicted arguments to match annotated spans exactly. However, this approach fails for generative models like large language models (LLMs), which produce diverse yet semantically accurate responses. EM underestimates performance by disregarding valid variations, implicit arguments (unstated but inferable), and scattered arguments (distributed across a document). To bridge this gap, we introduce Reliable Evaluation framework for Generative event argument extraction (REGen), a framework that better aligns with human judgment. Across six datasets, REGen improves performance by an average of 23.93 F1 points over EM. Human validation further confirms REGen's effectiveness, achieving 87.67% alignment with human assessments of argument correctness.
Assuming a directed acyclic graph (DAG) that represents prior knowledge of causal relationships between variables is a common starting point for cause-effect estimation. Existing literature typically invokes hypothetical domain expert knowledge or causal discovery algorithms to justify this assumption. In practice, neither may propose a single DAG with high confidence. Domain experts are hesitant to rule out dependencies with certainty or have ongoing disputes about relationships; causal discovery often relies on untestable assumptions itself or only provides an equivalence class of DAGs and is commonly sensitive to hyperparameter and threshold choices. We propose an efficient, gradient-based optimization method that provides bounds for causal queries over a collection of causal graphs -- compatible with imperfect prior knowledge -- that may still be too large for exhaustive enumeration. Our bounds achieve good coverage and sharpness for causal queries such as average treatment effects in linear and non-linear synthetic settings as well as on real-world data. Our approach aims at providing an easy-to-use and widely applicable rebuttal to the valid critique of `What if your assumed DAG is wrong?'.
The Distributed Constraint Optimization Problem (DCOP) formulation is a powerful tool to model cooperative multi-agent problems that need to be solved distributively. A core assumption of existing approaches is that DCOP solutions can be easily understood, accepted, and adopted, which may not hold, as evidenced by the large body of literature on Explainable AI. In this paper, we propose the Explainable DCOP (X-DCOP) model, which extends a DCOP to include its solution and a contrastive query for that solution. We formally define some key properties that contrastive explanations must satisfy for them to be considered as valid solutions to X-DCOPs as well as theoretical results on the existence of such valid explanations. To solve X-DCOPs, we propose a distributed framework as well as several optimizations and suboptimal variants to find valid explanations. We also include a human user study that showed that users, not surprisingly, prefer shorter explanations over longer ones. Our empirical evaluations showed that our approach can scale to large problems, and the different variants provide different options for trading off explanation lengths for smaller runtimes. Thus, our model and algorithmic contributions extend the state of the art by reducing the barrier for users to understand DCOP solutions, facilitating their adoption in more real-world applications.
A fundamental problem in combinatorial optimization is identifying equivalent formulations. Despite the growing need for automated equivalence checks -- driven, for example, by optimization copilots, which generate problem formulations from natural language descriptions -- current approaches rely on simple heuristics that fail to reliably check formulation equivalence. Inspired by Karp reductions, in this work we introduce Quasi-Karp equivalence, a formal criterion for determining when two optimization formulations are equivalent based on the existence of a mapping between their decision variables. We propose EquivaMap, a framework that leverages large language models to automatically discover such mappings for scalable, reliable equivalence checking, with a verification stage that ensures mapped solutions preserve feasibility and optimality without additional solver calls. To evaluate our approach, we construct EquivaFormulation, the first open-source dataset of equivalent optimization formulations, generated by applying transformations such as adding slack variables or valid inequalities to existing formulations. Empirically, EquivaMap significantly outperforms existing methods, achieving substantial improvements in correctly identifying formulation equivalence.
The assessment of argument quality depends on well-established logical, rhetorical, and dialectical properties that are unavoidably subjective: multiple valid assessments may exist, there is no unequivocal ground truth. This aligns with recent paths in machine learning, which embrace the co-existence of different perspectives. However, this potential remains largely unexplored in NLP research on argument quality. One crucial reason seems to be the yet unexplored availability of suitable datasets. We fill this gap by conducting a systematic review of argument quality datasets. We assign them to a multi-layered categorization targeting two aspects: (a) What has been annotated: we collect the quality dimensions covered in datasets and consolidate them in an overarching taxonomy, increasing dataset comparability and interoperability. (b) Who annotated: we survey what information is given about annotators, enabling perspectivist research and grounding our recommendations for future actions. To this end, we discuss datasets suitable for developing perspectivist models (i.e., those containing individual, non-aggregated annotations), and we showcase the importance of a controlled selection of annotators in a pilot study.
Conformal prediction provides a powerful framework for constructing prediction intervals with finite-sample guarantees, yet its robustness under distribution shifts remains a significant challenge. This paper addresses this limitation by modeling distribution shifts using Levy-Prokhorov (LP) ambiguity sets, which capture both local and global perturbations. We provide a self-contained overview of LP ambiguity sets and their connections to popular metrics such as Wasserstein and Total Variation. We show that the link between conformal prediction and LP ambiguity sets is a natural one: by propagating the LP ambiguity set through the scoring function, we reduce complex high-dimensional distribution shifts to manageable one-dimensional distribution shifts, enabling exact quantification of worst-case quantiles and coverage. Building on this analysis, we construct robust conformal prediction intervals that remain valid under distribution shifts, explicitly linking LP parameters to interval width and confidence levels. Experimental results on real-world datasets demonstrate the effectiveness of the proposed approach.
We consider the problem of conformal prediction under covariate shift. Given labeled data from a source domain and unlabeled data from a covariate shifted target domain, we seek to construct prediction sets with valid marginal coverage in the target domain. Most existing methods require estimating the unknown likelihood ratio function, which can be prohibitive for high-dimensional data such as images. To address this challenge, we introduce the likelihood ratio regularized quantile regression (LR-QR) algorithm, which combines the pinball loss with a novel choice of regularization in order to construct a threshold function without directly estimating the unknown likelihood ratio. We show that the LR-QR method has coverage at the desired level in the target domain, up to a small error term that we can control. Our proofs draw on a novel analysis of coverage via stability bounds from learning theory. Our experiments demonstrate that the LR-QR algorithm outperforms existing methods on high-dimensional prediction tasks, including a regression task for the Communities and Crime dataset, and an image classification task from the WILDS repository.