Ziyi Wang, Yuxuan Lu, Wenbo Li, Amirali Amini, Bo Sun, Yakov Bart, Weimin Lyu, Jiri Gesi, Tian Wang, Jing Huang, Yu Su, Upol Ehsan, Malihe Alikhani, Toby Jia-Jun Li, Lydia Chilton, Dakuo Wang. Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2026.

Detecting deviant language such as sexism, or nuanced language such as metaphors or sarcasm, is crucial for enhancing the safety, clarity, and interpretation of social interactions. While existing classifiers deliver strong results on these tasks, they often come with significant computational cost and high data demands. In this work, we propose Class Distillation (ClaD), a novel training paradigm that targets the core challenge: distilling a small, well-defined target class from a highly diverse and heterogeneous background. ClaD integrates two key innovations: (i) a loss function informed by the structural properties of class distributions, based on Mahalanobis distance, and (ii) an interpretable decision algorithm optimized for class separation. Across three benchmark detection tasks – sexism, metaphor, and sarcasm – ClaD outperforms competitive baselines, and even with smaller language models and orders of magnitude fewer parameters, achieves performance comparable to several large language models. These results demonstrate ClaD as an efficient tool for pragmatic language understanding tasks that require gleaning a small target class from a larger heterogeneous background.

Crime prediction (CP) plays a pivotal role in urban analytics, contributing significantly to personal safety and societal stability. Unlike conventional time series forecasting, CP faces unique difficulties due to the inherent sparsity of crime incidents, particularly within small spatial regions and limited time windows. This sparsity, coupled with the non-Gaussian distribution of crime data—characterized by an excess of zero events and over-dispersion—presents a critical challenge for the signal processing community. In this regard, we propose a novel framework, Spatial-Temporal Multivariate Zero-Inflated Negative Binomial Graph Neural Networks (STMGNN-ZINB), which integrates diffusion and convolutional graph networks to capture spatial, temporal, and multivariate dependencies. By leveraging a Zero-Inflated Negative Binomial distribution, the STMGNN-ZINB effectively models the over-dispersed and zero-heavy nature of crime data, significantly improving both prediction accuracy and confidence interval estimation. Experimental results on real-world datasets demonstrate that our STMGNN-ZINB outperforms state-of-the-art CP methods, offering a robust tool for crime early warning and explicable insights into urban crime dynamics.

Handling implicit language is essential for natural language processing systems to achieve precise text understanding and facilitate natural interactions with users. Despite its importance, the absence of a metric for accurately measuring the implicitness of language significantly constrains the depth of analysis possible in evaluating models' comprehension capabilities. This paper addresses this gap by developing a scalar metric that quantifies the implicitness level of language without relying on external references. Drawing on principles from traditional linguistics, we define "implicitness" as the divergence between semantic meaning and pragmatic interpretation. To operationalize this definition, we introduce ImpScore, a reference-free metric formulated through an interpretable regression model. This model is trained using pairwise contrastive learning on a specially curated dataset consisting of (*implicit sentence*, *explicit sentence*) pairs. We validate ImpScore through a user study that compares its assessments with human evaluations on out-of-distribution data, demonstrating its accuracy and strong correlation with human judgments. Additionally, we apply ImpScore to hate speech detection datasets, illustrating its utility and highlighting significant limitations in current large language models' ability to understand highly implicit content. Our metric is publicly available at https://github.com/audreycs/ImpScore.

The emergence of Vision-Language Models (VLMs) represents a significant advancement in integrating computer vision with Large Language Models (LLMs) to generate detailed text descriptions from visual inputs. Despite their growing importance, the security of VLMs, particularly against backdoor attacks, is under explored. Moreover, prior works often assume attackers have access to the original training data, which is often unrealistic. In this paper, we address a more practical and challenging scenario where attackers must rely solely on Out-Of-Distribution (OOD) data. We introduce VLOOD (Backdoor Vision-Language Models using Out-of-Distribution Data), a novel approach with two key contributions: (1) demonstrating backdoor attacks on VLMs in complex image-to-text tasks while minimizing degradation of the original semantics under poisoned inputs, and (2) proposing innovative techniques for backdoor injection without requiring any access to the original training data. Our evaluation on image captioning and visual question answering (VQA) tasks confirms the effectiveness of VLOOD, revealing a critical security vulnerability in VLMs and laying the foundation for future research on securing multimodal models against sophisticated threats.

Deep learning has achieved significant success by training on balanced datasets. However, real-world data often exhibit long-tailed distributions. Empirical studies have revealed that long-tailed data skew data representations, where head classes dominate the feature space. Many methods have been proposed to empirically rectify the skewed representations. However, a clear understanding of the underlying cause and extent of this skew remains lacking. In this study, we provide a comprehensive theoretical analysis to elucidate how long-tailed data affect feature distributions, deriving the conditions under which centers of tail classes shrink together or even collapse into a single point. This results in overlapping feature distributions of tail classes, making features in the overlapping regions inseparable. Moreover, we demonstrate that merely empirically correcting the skewed representations of the training data is insufficient to separate the overlapping features due to distribution shifts between the training and real data. To address these challenges, we propose a novel long-tailed representation learning method, FeatRecon. It reconstructs the feature space in order to arrange features from different classes into symmetricial and linearly separable regions. This, in turn, enhances the model’s robustness to long-tailed data. We validate the effectiveness of our method through extensive experiments on the CIFAR-10-LT, CIFAR-100-LT, ImageNet-LT, and iNaturalist 2018 datasets.

Concept Bottleneck Models (CBMs) have garnered much attention for their ability to elucidate the prediction process through a human-understandable concept layer. However, most previous studies focused on cases where the data, including concepts, are clean. In many scenarios, we always need to remove/insert some training data or new concepts from trained CBMs due to different reasons, such as privacy concerns, data mislabelling, spurious concepts, and concept annotation errors. Thus, the challenge of deriving efficient editable CBMs without retraining from scratch persists, particularly in large-scale applications. To address these challenges, we propose Editable Concept Bottleneck Models (ECBMs). Specifically, ECBMs support three different levels of data removal: concept-label-level, concept-level, and data-level. ECBMs enjoy mathematically rigorous closed-form approximations derived from influence functions that obviate the need for re-training. Experimental results demonstrate the efficiency and effectiveness of our ECBMs, affirming their adaptability within the realm of CBMs.

Self-supervised learning plays an important role in current state-of-the-art semi-supervised learning (SSL) methods. These methods learn inter-class heterogeneity among data and generate pseudo-labels based on class level representations. However, they often neglect intra-class heterogeneity, resulting in the under-exploitation of finer-grained semantic relationships within classes. To address this limitation, we introduce PivotAlign, a novel SSL approach that aims to 1) learn hierarchical representations to detect both interclass and intra-class semantic relationships, and 2) refine pseudo-labels based on learned representations with a class-debiasing strategy. Specifically, we first learn a set of pivots as sub-prototypes of classes. We then train representations so that features align with the assigned pivot and are hierarchically grouped based on both inter-class and intra-class heterogeneity. This allows us to capture both inter-class and intra-class semantic relationships among data and leverage them to better assign and refine pseudo-labels. Additionally, since SSL methods are prone to bias toward classes that are easier to learn, we further re-balance class predictions to alleviate this class bias. We demonstrate the effectiveness of PivotAlign on various SSL benchmarks, where PivotAlign achieves state-of-the-art performances. The source code will be released upon publication of the work.

Manufacturing is entering a new era with Industrial Internet and edge computing. The Industrial Internet cloud platform provides massive computing power and storage spaces for field devices. With more powerful chips available, field devices are also capable of handling multiple complex computational tasks simultaneously. How to collaborate resources from both cloud platforms and edge devices become an important topic for manufacturers. In this paper, a cloud-edge collaborated batch control system is proposed based on the IEC 61499 standard and the ISA-88 standard. The ISA-88 models are implemented as an independent IEC 61499 resource to support low-code development for batch control systems. Also, cloud resources are introduced in the IEC 61499 deployment to enable cloud-edge collaboration. Finally, the design process is verified with a liquid food processing line.

Trojan attacks raise serious security concerns.In this paper, we investigate the underlying mechanism of Trojaned BERT models.We observe the attention focus drifting behavior of Trojaned models, i.e., when encountering an poisoned input, the trigger token hijacks the attention focus regardless of the context.We provide a thorough qualitative and quantitative analysis of this phenomenon, revealing insights into the Trojan mechanism.Based on the observation, we propose an attention-based Trojan detector to distinguish Trojaned models from clean ones.To the best of our knowledge, this is the first paper to analyze the Trojan mechanism and to develop a Trojan detector based on the transformer's attention 1 .