He Geng, Yangmin Huang, Lixian Lai, Qianyun Du, Hui Chu, Zhiyang He, Jiaxue Hu, Xiaodong Tao. Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2026.

The integration of multi-omics data, including DNA methylation, mRNA expression, and miRNA profiles, is crucial for accurate disease subtyping and outcome prediction in complex disorders such as Alzheimer's disease and various cancers. However, the inherent heterogeneity and inconsistency among omics views present significant challenges for reliable data fusion. To address these issues, we propose KANTrust, a novel framework for trustworthy multi-omics classification that explicitly models both epistemic and aleatoric uncertainties. Our method combines a Kolmogorov-Arnold Network (KAN)enhanced robust representation module, a contrastive evidence consistency module, and an evidence-theoretic fusion module to achieve reliable multi-view integration. KANTrust adaptively highlights informative features within each omics modality, promotes semantic alignment across views, and quantifies uncertainty through a Dempster-Shafer framework. Experimental evaluations on four real-world biomedical datasets demonstrate that KANTrust consistently outperforms state-of-the-art methods in both binary and multi-class classification tasks. Code is available at https://github.com/wcj6/KANTrust.

The concept bottleneck model (CBM), as a technique improving interpretability via linking predictions to human-understandable concepts, makes high-risk and life-critical medical image classification credible. Typically, existing CBM methods associate the final layer of visual encoders with concepts to explain the model’s predictions. However, we empirically discover the phenomenon of concept preference variation, that is, the concepts are preferably associated with the features at different layers than those only at the final layer; yet a blind last-layer-based association neglects such a preference variation and thus weakens the accurate correspondences between features and concepts, impairing model interpretability. To address this issue, we propose a novel Multi-layer Visual Preference-enhanced Concept Bottleneck Model (MVP-CBM), which comprises two key novel modules: (1) intra-layer concept preference modeling, which captures the preferred association of different concepts with features at various visual layers, and (2) multi-layer concept sparse activation fusion, which sparsely aggregates concept activations from multiple layers to enhance performance. Thus, by explicitly modeling concept preferences, MVP-CBM can comprehensively leverage multi-layer visual information to provide a more nuanced and accurate explanation of model decisions. Extensive experiments on several public medical classification benchmarks demonstrate that MVP-CBM achieves state-of-the-art accuracy and interoperability, verifying its superiority. Code is available at https://github.com/wcj6/MVP-CBM.

The advancement of Zero-Shot Learning in the medi-cal domain has been driven forward by using pretrained models on large-scale image-text pairs, focusing on image-text alignment. However, existing methods primarily rely on cosine similarity for alignment, which may not fully capture the complex relationship between medical images and reports. To address this gap, we introduce a novel approach called Cross-Attention Alignment for Radiology Zero-Shot Classification (CARZero). Our approach innovatively leverages cross-attention mechanisms to process image and report features, creating a Similarity Representation that more accurately reflects the intricate relationships in medical semantics. This representation is then linearly projected to form an image-text similarity matrix for cross-modality alignment. Additionally, recognizing the pivotal role of prompt selection in zero-shot learning, CARZero in-corporates a Large Language Model-based prompt alignment strategy. This strategy standardizes diverse diagnostic expressions into a unified format for both training and inference phases, overcoming the challenges of manual prompt design. Our approach is simple yet effective, demonstrating state-of-the-art performance in zero-shot classification on five official chest radiograph diagnostic test sets, including remarkable results on datasets with long-tail distributions of rare diseases. This achievement is attributed to our new image-text alignment strategy, which effectively addresses the complex relationship between medical images and reports. Code and models are available at https://github.com/laihaoran/CARZero.

In the minimum k-cut problem, we want to find the minimum number of edges whose deletion breaks the input graph into at least k connected components. The classic algorithm of Karger and Stein runs in Õ(n2k−2) time, and recent, exciting developments have improved the running time to O(nk). For general, weighted graphs, this is tight assuming popular hardness conjectures.

Let $G$ be a graph and $S, T \subseteq V(G)$ be (possibly overlapping) sets of terminals, $|S|=|T|=k$. We are interested in computing a vertex sparsifier for terminal cuts in $G$, i.e., a graph $H$ on a smallest possible number of vertices, where $S \cup T \subseteq V(H)$ and such that for every $A \subseteq S$ and $B \subseteq T$ the size of a minimum $(A,B)$-vertex cut is the same in $G$ as in $H$. We assume that our graphs are unweighted and that terminals may be part of the min-cut. In previous work, Kratsch and Wahlström (FOCS 2012/JACM 2020) used connections to matroid theory to show that a vertex sparsifier $H$ with $O(k^3)$ vertices can be computed in randomized polynomial time, even for arbitrary digraphs $G$. However, since then, no improvements on the size $O(k^3)$ have been shown. In this paper, we draw inspiration from the renowned Bollobás's Two-Families Theorem in extremal combinatorics and introduce the use of total orderings into Kratsch and Wahlström's methods. This new perspective allows us to construct a sparsifier $H$ of $Θ(k^2)$ vertices for the case that $G$ is a DAG. We also show how to compute $H$ in time near-linear in the size of $G$, improving on the previous $O(n^{ω+1})$. Furthermore, $H$ recovers the closest min-cut in $G$ for every partition $(A,B)$, which was not previously known. Finally, we show that a sparsifier of size $Ω(k^2)$ is required, both for DAGs and for undirected edge cuts.

In this paper we study the maximum number of hyperedges which may be in an $r$-uniform hypergraph under the restriction that no pair of vertices has more than $t$ Berge paths of length $k$ between them. When $r=t=2$, this is the even-cycle problem asking for ${ex}(n, C_{2k})$. We extend results of Füredi and Simonovits and of Conlon, who studied the problem when $r=2$. In particular, we show that for fixed $k$ and $r$, there is a constant $t$ such that the maximum number of edges can be determined in order of magnitude.

Reading and understanding text is one important component in computer aided diagnosis in clinical medicine, also being a major research problem in the field of NLP. In this work, we introduce a question-answering task called MedQA to study answering questions in clinical medicine using knowledge in a large-scale document collection. The aim of MedQA is to answer real-world questions with large-scale reading comprehension. We propose our solution SeaReader---a modular end-to-end reading comprehension model based on LSTM networks and dual-path attention architecture. The novel dual-path attention models information flow from two perspectives and has the ability to simultaneously read individual documents and integrate information across multiple documents. In experiments our SeaReader achieved a large increase in accuracy on MedQA over competing models. Additionally, we develop a series of novel techniques to demonstrate the interpretation of the question answering process in SeaReader.

In the current paper, we propose a probabilistic generative model, the label correlation mixture model (LCMM), to depict multi-labeled document data, which can be utilized for multi-label text classification. LCMM assumes two stochastic generative processes, which correspond to two submodels: 1) a label correlation model; and 2) a label mixture model. The former model formulates labels’ generative process, in which a label correlation network is created to depict the dependency between labels. Moreover, we present an efficient inference algorithm for calculating the generative probability of a multi-label class. Furthermore, in order to optimize the label correlation network, we propose a parameter-learning algorithm based on gradient descent. The second submodel in the LCMM depicts the generative process of words in a document with the given labels. Different traditional mixture models can be adopted in this generative process, such as the mixture of language models, or topic models. In the multi-label classification stage, we propose a two-step strategy to most efficiently utilize the LCMM based on the framework of Bayes decision theory. We conduct extensive multi-label classification experiments on three standard text data sets. The experimental results show significant performance improvements comparing to existing approaches. For example, the improvements on accuracy and macro F-score measures in the OHSUMED data set achieve 28.3% and 37.0%, respectively. These performance enhancements demonstrate the effectiveness of the proposed models and solutions.

We propose a new unsupervised learning model, hidden softmax sequence model (HSSM), based on Boltzmann machine for dialogue structure analysis.The model employs three types of units in the hidden layer to discovery dialogue latent structures: softmax units which represent latent states of utterances; binary units which represent latent topics specified by dialogues; and a binary unit that represents the global general topic shared across the whole dialogue corpus.In addition, the model contains extra connections between adjacent hidden softmax units to formulate the dependency between latent states.Two different kinds of real world dialogue corpora, Twitter-Post and AirTicketBooking, are utilized for extensive comparing experiments, and the results illustrate that the proposed model outperforms sate-ofthe-art popular approaches.

Multi-label text categorization is more difficult but practical than the conventional binary or multi-class text categorization. This paper propose a novel probabilistic generative model, label correlation mixture model (LCMM), to depict the multiple labeled documents, which can be used for multi-label text categorization. In LCMM, labels and topics have the one-to-one correspondences. LCMM consists of two parts: label correlation model and multi-label conditioned document model. The former one formulates the generating process of labels and the dependencies between the labels are taken into account. We also propose an efficient algorithm for calculating the probability of generating an arbitrary subset of labels. Multi-label conditioned document model can be regarded as a supervised label mixture model, in which the labels for a document are known. To evaluate LCMM, multi-label text categorization experiments on three standard text data sets are performed. The experimental results demonstrate the effectiveness of LCMM, comparing to other reported methods.