Deep multi-view clustering (MVC) methods achieve impressive performance by effectively capturing complementary information across views, where feature fusion serves as the critical mechanism for maximizing cross-view complementarity. However, most existing methods suffer from rigid dependence on non-adaptive predefined fusion operations, resulting in unverifiable and potentially suboptimal fused feature quality. To resolve these limitations, we propose a novel multi-view clustering framework that learns adaptive hierarchical fusion through an unsupervised evolutionary algorithm. Unlike conventional predefined-fusion strategies, our approach employs tree-structured representations (Fusion Trees) for adaptive feature integration. These Fusion Trees are optimized via our evolutionary mechanism, in which models sharing identical architectures but distinct Fusion Trees are conceptualized as evolutionary individuals. Through implementation of the evolutionarily optimized Fusion Tree, the resultant model generates discriminative representations in accordance with biological evolutionary principles. Comprehensive benchmarking across twelve multi-view datasets validates significant performance gains improvement over state-of-the-art baselines.

Yue Fan, Hu Zhang, Yunxiao Zhao, Guangjun Zhang, Hao Zhan, Ru Li, Hongye Tan, Yuanlong Wang. Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2026.

Ya Su, Hu Zhang, Dan Qiao, YuJie Wang, Yunxiao Zhao, Yue Fan, Shike Li, Ru Li, Hongye Tan. Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2026.

Rationalization, a data-centric framework, aims to build self-explanatory models to explain the prediction outcome by generating a subset of human-intelligible pieces of the input data. It involves a cooperative game model where a generator generates the most human-intelligible parts of the input (i.e., rationales), followed by a predictor that makes predictions based on these generated rationales. Conventional rationalization methods typically impose constraints via regularization terms to calibrate or penalize undesired generation. However, these methods are suffering from a problem called mode collapse, in which the predictor produces correct predictions yet the generator consistently outputs rationales with collapsed patterns. Moreover, existing studies are typically designed separately for specific collapsed patterns, lacking a unified consideration. In this paper, we systematically revisit cooperative rationalization from a novel game-theoretic perspective and identify the fundamental cause of this problem: the generator no longer tends to explore new strategies to uncover informative rationales, ultimately leading the system to converge to a suboptimal game equilibrium (correct predictionsv.scollapsed rationales). To solve this problem, we then propose a novel approach, Game-theoreticPolicyOptimization orientedRATionalization (PoRat), which progressively introduces policy interventions to address the game equilibrium in the cooperative game process, thereby guiding the model toward a more optimal solution state. We theoretically analyse the cause of such a suboptimal equilibrium and prove the feasibility of the proposed method. Furthermore, we validate our method on nine widely used real-world datasets and two synthetic settings, wherePoRatachieves up to 8.1% performance improvements over existing state-of-the-art methods. Code and data are available atanonymous.4open.science/r/Rationalization-PORAT-ECE9.

Multi-hop question answering (MHQA) aims to utilize multi-source intensive documents retrieved to derive the answer. However, it is very challenging to model the importance of knowledge retrieved. Previous approaches primarily emphasize single-step and multi-step iterative decomposition or retrieval, which are susceptible to failure in long-chain reasoning due to the progressive accumulation of erroneous information. To address this problem, we propose a novel Local-tO-Global optimized retrieval method (LOG) to discover more beneficial information, facilitating the MHQA. In particular, we design a pointwise conditional v-information based local information modeling to cover usable documents with reasoning knowledge. We also improve tuplet objective loss, advancing multi-examples-aware global optimization to model the relationship between scattered documents. Extensive experimental results demonstrate our proposed method outperforms prior state-of-the-art models, and it can significantly improve multi-hop reasoning, notably for long-chain reasoning.

The Industrial Internet of Things (IIoT) represents a sophisticated technology designed to enhance production management and predict output in industrial settings, including machinery fault diagnostics. The precision of fault diagnosis is contingent upon the training efficacy of diagnostic models and their interoperability with models from other industrial facilities. Nonetheless, several critical challenges persist in maintaining these diagnostic models: 1) machinery sensors may generate abnormal data, resulting in suboptimal quality in model training; 2) sensor malfunctions may lead to interruptions in continuous data flow, thus impeding model training; and 3) collaborative interactions with other factories aiming at improving model performance may pose risks of privacy breaches. In this study, we introduce the FCLLM-DT scheme, which integrates the digital twin (DT) methodology to create a physical model of bearing for fixing abnormal sensor data. Additionally, retrieval-augmented generation (RAG)-assisted large language models (LLMs) are utilized to generate virtual datasets in instances of sensor failure. Moreover, for IIoT applications across distributed industrial environments, federated continual learning (FCL) is employed to enhance global model training by aggregating localized models from diverse facilities, thereby improving the accuracy of bearing fault diagnosis while safeguarding data privacy. The experiments on the accuracy of DT for abnormal data fix, RAG-assisted LLM for virtual data generation, and FCL for bearing fault diagnosis are conducted in comparison with three alternative methods across two datasets. The results indicate that our proposed scheme surpasses existing methods in both the enhancement of sensing data quality and the accuracy of bearing fault diagnosis.

As a leading graph clustering technique, spectral clustering is one of the most widely used clustering methods to capture complex clusters in data. Some additional prior information can help it to further reduce the difference between its clustering results and users' expectations. However, it is hard to get the prior information under unsupervised scene to guide the clustering process. To solve this problem, we propose a self-constrained spectral clustering algorithm. In this algorithm, we extend the objective function of spectral clustering by adding pairwise and label self-constrained terms to it. We provide the theoretical analysis to show the roles of the self-constrained terms and the extensibility of the proposed algorithm. Based on the new objective function, we build an optimization model for self-constrained spectral clustering so that we can simultaneously learn the clustering results and constraints. Furthermore, we propose an iterative method to solve the new optimization problem. Compared to other existing versions of spectral clustering algorithms, the new algorithm can discover a high-quality cluster structure of a data set without prior information. Extensive experiments on benchmark data sets illustrate the effectiveness of the proposed algorithm.