Recent research aims to improve cross-lingual transfer learning in low-resource scenarios by optimizing the internal representations of multilingual models. However, previous methods typically rely on large-scale parallel corpora and overlook the subtle semantic differences between languages, leading to excessive clustering or collapse of local semantic units in the embedding space. This weakens the model's generalization ability on unseen data and its robustness under noisy conditions. To address this, we propose XLingLearn, a collaborative optimization framework that enhances cross-lingual transfer by expanding robust embedding regions and refining multilingual embedding space. Specifically, we design a distance-dispersion constraint strategy to push overly clustered semantic units apart to prevent semantic collapse, and introduce a direction-consistency constraint to prevent semantic bias. Additionally, we introduce an attention consistency module to stabilize the robust region. Finally, to mitigate the embedding space and context space mismatch caused by data augmentation, we introduce a debiasing-optimization regularization term, which enhances transfer efficiency and stability. Experimental results show that XLingLearn improves cross-lingual transfer performance across 17 target languages in the XNLI and PAWS-X tasks, enhancing generalization and robustness under low-resource and non-parallel conditions.

Positive-unlabeled (PU) learning handles the problem of learning a predictive model from PU data. Past few years have witnessed the boom of PU learning, while the existing learning algorithms are limited to binary classification and cannot be directly applied to multi-class PU data. In this paper, we present an unbiased estimator of the original classification risk for multi-class PU learning, and show that the direct empirical risk minimization suffers from the severe overfitting problem because the risk is unbounded below. To address this problem, we propose an alternative risk estimator, and theoretically establish an estimation error bound. We show that the estimation error of its empirical risk minimizer achieves the optimal parametric convergence rate. Extensive experiments on multiple datasets demonstrate the effectiveness of the proposed approach for multi-class PU learning.

Partial multi-label learning (PML) deals with the problem where each training example is assigned multiple candidate labels, only a part of which are correct. To learn from such PML examples, the straightforward model training tends to be misled by the noise candidate label set. To alleviate this problem, a coupled framework is established in this paper to learn the desired model and perform the relabeling procedure alternatively. In the relabeling procedure, instead of simply extracting relative label confidences, or deterministically eliminating low confidence labels and preserving high confidence labels as ground-truth ones, we introduce a soft sign thresholding operator to adaptively strengthen candidate labels with high confidence and weaken candidate labels with low confidence, which enlarges the difference of confidences of candidate labels within allowable range. We further show that the resulting nonconvex quadratic programming (QP) optimization problem can be relaxed into a convex QP problem with proper conditions. Extensive experiments on synthesized and real-world data sets demonstrate the effectiveness of our proposed approach.

Partial label learning aims to induce a multi-class classifier from training examples where each of them is associated with a set of candidate labels, among which only one is the ground-truth label. The common strategy to train predictive model is disambiguation, i.e. differentiating the modeling outputs of individual candidate labels so as to recover ground-truth labeling information. Recently, feature-aware disambiguation was proposed to generate different labeling confidences over candidate label set by utilizing the graph structure of feature space. However, the existence of noise and outliers in training data makes the similarity derived from original features less reliable. To this end, we proposed a novel approach for partial label learning based on adaptive graph guided disambiguation (PL-AGGD). Compared with fixed graph, adaptive graph could be more robust and accurate to reveal the intrinsic manifold structure within the data. Moreover, instead of the two-stage strategy in previous algorithms, our approach performs label disambiguation and predictive model training simultaneously. Specifically, we present a unified framework which jointly optimizes the ground-truth labeling confidences, similarity graph and model parameters to achieve strong generalization performance. Extensive experiments show that PL-AGGD performs favorably against state-of-the-art partial label learning approaches.

In multi-label learning, each instance is associated with multiple labels simultaneously. Most of the existing approaches directly treat each label in a crisp manner, i.e. one class label is either relevant or irrelevant to the instance. However, the latent relative importance of each relevant label is regrettably ignored. In this paper, we propose a novel multi-label learning approach that aims to estimate the latent labeling importances while training the inductive model simultaneously. Specifically, we present a biconvex formulation with both instance and label graph regularization, and solve this problem using an alternating way. On the one hand, the inductive model is trained by minimizing the least squares loss of fitting the latent relative labeling importances. On the other hand, the latent relative labeling importances are estimated by the modeling outputs via a specially constrained label propagation procedure. Through the mutual adaption of the inductive model training and the specially constrained label propagation, an effective multi-label learning model is therefore built by optimally estimating the latent relative labeling importances. Extensive experimental results clearly show the effectiveness of the proposed approach.

Recommender systems play an important role in providing personalized information to users and helping address the information overload problem. Recent research has considered social theories and studied the importance of social influence in social recommendation systems. However, many publications ignored the users' roles information or just considered some single roles. In fact, users often have many different roles. Besides, different types of users (users with different roles) might have different conformity tendency. Thus, this inspires us to study how conformity tendency changes with users' roles in recommender systems. We firstly formalize conformity influence by defining a utility function and then propose a probabilistic graphical model integrating both users' roles and conformity tendency, named as Role Conformity Recommender Systems (RCRS). We evaluate the proposed model on several real-world datasets. The experimental results show that our model significantly outperforms state-of-the-art approaches.