Out-of-distribution (OOD) detection is a well-known challenge due to deep models often producing overconfident. In this paper, we reveal a key insight that trained classifiers tend to rely on sparse parameter contribution patterns, meaning that only a few dominant parameters drive predictions. This brittleness can be exploited by OOD inputs that anomalously trigger these parameters, resulting in overconfident predictions. To address this issue, we propose a simple yet effective method called Shaping Parameter Contribution Patterns (SPCP), which enhances OOD detection robustness by encouraging the classifier to learn boundary-oriented dense contribution patterns. Specifically, SPCP operates during training by rectifying excessively high parameter contributions based on a dynamically estimated threshold. This mechanism promotes the classifier to rely on a broader set of parameters for decision-making, thereby reducing the risk of overconfident predictions caused by anomalously triggered parameters, while preserving in-distribution (ID) performance. Extensive experiments under various OOD detection setups verify the effectiveness of SPCP.

Xiangyu Wu, Yuwei Hu, Tianyu Cui, Yueying Tian, Qing-Guo Chen, Zhao Xu, Weihua Luo, Kaifu Zhang, Yang Yang, Jianfeng Lu. Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2026.

Out-of-distribution (OOD) detection is crucial for ensuring the reliable deployment of deep models in real-world scenarios. Recently, from the perspective of over-parameterization, a series of methods leveraging weight sparsification techniques have shown promising performance. These methods typically focus on selecting important parameters for in-distribution (ID) data to reduce the negative impact of redundant parameters on OOD detection. However, we empirically find that these selected parameters may behave overconfidently toward OOD data and hurt OOD detection. To address this issue, we propose a simple yet effective post-hoc method called Instance-aware Test Pruning (ITP), which performs OOD detection by considering both coarse-grained and fine-grained levels of parameter pruning. Specifically, ITP first estimates the class-specific parameter contribution distribution by exploring the ID data. By using the contribution distribution, ITP conducts coarse-grained pruning to eliminate redundant parameters. More importantly, ITP further adopts a fine-grained test pruning process based on the right-tailed Z-score test, which can adaptively remove instance-level overconfident parameters. Finally, ITP derives OOD scores from the pruned model to achieve more reliable predictions. Extensive experiments on widely adopted benchmarks verify the effectiveness of ITP, demonstrating its competitive performance.

Replay-based class incremental learning (CIL) mitigates catastrophic forgetting by retaining limited historical data. However, the class imbalance between historical and new class data prevents catastrophic forgetting from being fully resolved. Existing approaches rely solely on buffer data to approximate the historical distribution, neglecting the incomplete historical distribution modeling caused by feature drift in the evolving feature space, limiting their effectiveness. To address this challenge, we propose a novel Feature Drift oriented Distribution Reconstruction (FDDR) framework, which reconstructs complete historical distributions utilizing both historical and new knowledge. Specifically, FDDR mitigates the structure degradation of historical distributions caused by feature drift by aligning distributions between the current and previous feature spaces, reserving sufficient feature spaces for subsequent distribution reconstruction. Based on the preserved space, FDDR supplements the sparse historical distribution by incorporating weighted new-class distributions according to inter-class relationships, which are derived from joint prototype cosine similarities. In particular, each joint prototype is constructed from both evolved historical prototypes and buffer prototypes, thereby yielding a more reliable similarity measure. Finally, building upon the reconstructed distributions, FDDR generates pseudo-features to correct biased classifier and further fine-tunes the feature space to achieve superior feature space reconstruction. Comprehensive experiments on widely adopted CIL benchmarks verify the effectiveness of FDDR. Code is available at https://github.com/njustkmg/ECAI25-FDDR.

With the development of foundation models, advanced research explores the potential of vision-language models (VLMs) for out-of-distribution (OOD) detection. In particular, methods that generate outliers for regularized prompt learning have shown promising results in few-shot settings. However, existing methods that rely solely on visual modalities struggle to synthesize outliers semantically analogous to in-distribution (ID) data, neglecting near OOD scenarios. Notably, recent large language models (LLMs) exhibit a deep real-world understanding, enabling the generation of near textual outliers. Inspired by this, we propose an OOD detection framework named TextAugmented Cues (TAC), which integrates expert knowledge from LLMs into the prompt learning of VLMs. Specifically, we first design LLMs query templates to generate outlier categories based on Visual Similarity. Then, LLMs are further leveraged to synthesize the semantic representations of ID and outlier categories based on Feature Distinctiveness. Subsequently, we incorporate visual-textual information of ID categories for prompt learning, regularized by textual outliers. Experimental results demonstrate that TAC significantly outperforms state-of-the-art (SOTA) VLMbased OOD detection methods in few-shot scenarios. The code is available at https://github.com/njustkmg/ICDM25-TAC.

Mainstream test-time adaptation (TTA) techniques endeavor to mitigate distribution shifts via entropy minimization for multi-class classification, inherently increasing the probability of the most confident class. However, when encountering multi-label instances, the primary challenge stems from the varying number of labels per image, and prioritizing only the highest probability class inevitably undermines the adaptation of other positive labels. To address this issue, we investigate TTA within multi-label scenario (ML--TTA), developing Bound Entropy Minimization (BEM) objective to simultaneously increase the confidence of multiple top predicted labels. Specifically, to determine the number of labels for each augmented view, we retrieve a paired caption with yielded textual labels for that view. These labels are allocated to both the view and caption, called weak label set and strong label set with the same size k. Following this, the proposed BEM considers the highest top-k predicted labels from view and caption as a single entity, respectively, learning both view and caption prompts concurrently. By binding top-k predicted labels, BEM overcomes the limitation of vanilla entropy minimization, which exclusively optimizes the most confident class. Across the MSCOCO, VOC, and NUSWIDE multi-label datasets, our ML--TTA framework equipped with BEM exhibits superior performance compared to the latest SOTA methods, across various model architectures, prompt initialization, and varying label scenarios. The code is available at https://github.com/Jinx630/ML-TTA.

Existing multi-modal learning methods often struggle with modality imbalance in real-world applications, leading to suboptimal performance. A core reason is that under the multi-modal joint learning paradigm, the dominant modality tends to control the learning process, leaving non-dominant modalities insufficiently learned and underutilized. Some attempts naturally introduce additional uni-modal objectives to alleviate this issue. However, they overlook the problem of optimization conflicts between multi-modal objectives and uni-modal objectives, which adversely impact overall performance. To address this challenge, we propose Multi-modal learning with Coordinative Uni-modal assisTance (MCUT), which incorporates collaborative uni-modal tasks alongside multi-modal tasks. Specifically, it leverages a meta-optimization approach to maximize the inner product of gradients originating from both tasks, alleviating optimization conflicts. Moreover, based on gradient fusion analysis, we enhance the assistance of uni-modal tasks by weighting the phases of meta-optimization, further boosting performance. Experiments on multiple multi-modal datasets demonstrate MCUT’s superiority over existing methods. The code is available in https://github.com/njustkmg/ICME25-MCUT.

Out-of-distribution (OOD) detection aims to ensure AI system reliability by rejecting inputs outside the training distribution. Recent work shows that memorizing atypical samples during later stages of training can hurt OOD detection, while strategies for forgetting them show promising improvements. However, directly forgetting atypical samples sacrifices ID generalization and limits the model’s OOD detection capability. To address this issue, we propose Progressive Self-Knowledge Distillation (PSKD) framework, which strengthens the OOD detection capability by leveraging self-provided uncertainty-embedded targets. Specifically, PSKD adaptively selects a self-teacher model from the training history using pseudo-outliers, facilitating the learning of uncertainty knowledge via multi-level distillation applied to features and responses. As a result, PSKD achieves better ID generalization and uncertainty estimation for OOD detection. Moreover, PSKD is orthogonal to most existing methods and can be integrated as a plugin to collaborate with them. Experimental results from multiple OOD scenarios verify the effectiveness and general applicability of PSKD.

Incremental learning (IL) aims to sequentially learn new tasks while mitigating catastrophic forgetting. Among various IL strategies, parameter-isolation methods stand out by using mask techniques to allocate distinct parameters to each task, explicitly addressing forgetting. However, existing approaches often disregard parameter dependencies, resulting in an over-reliance on newly allocated parameters. To address this issue, we propose Probabilistic Group Mask selection (PGM), a group-wise approach that captures parameter dependencies by exploring candidate masks within each group. Specifically, PGM partitions parameters into groups with multiple candidate masks, assigning probabilities to these masks and leveraging Gumbel-Softmax for differentiable sampling, enabling efficient optimization of the discrete mask selection process. Our theoretical analysis demonstrates that incorporating parameter dependencies enhances sub-network selection. Experiments conducted on standard benchmarks confirm its superior effectiveness compared to existing IL approaches. The source code is available at: \url{https://github.com/njustkmg/ICML25-PGM}.

Multi-modal learning (MML) is frequently hindered by modality imbalance, leading to suboptimal performance in real-world applications. To address this issue, existing approaches primarily focus on rebalancing MML from the perspective of optimization or architecture design. However, almost all existing methods ignore the impact of sample sequences, i.e., an inappropriate training order tends to trigger learning bias in the model, further exacerbating modality imbalance. In this paper, we propose Balance-aware Sequence Sampling (BSS) to enhance the robustness of MML. Specifically, we first define a multi-perspective measurer to evaluate the balance degree of each sample in terms of correlation and information criteria. Via this evaluation, we employ a heuristic scheduler based on curriculum learning (CL) that incrementally provides training subsets, progressing from balanced to imbalanced samples to alleviate the imbalance. Moreover, we propose a learning-based probabilistic sampling method to dynamically update the training sequence in a more fine-grained manner, further improving MML performance. Extensive experiments on widely used datasets demonstrate the superiority of our method compared with state-of-the-art (SOTA) baselines. The code is available at https://github.com/njustkmg/IJCAI25-BSS.

Due to the notorious modality imbalance phenomenon, multimodal learning (MML) struggles to achieve satisfactory performance. Recently, multimodal learning with alternating unimodal adaptation (MLA) has been proven effective in mitigating the interference between modalities by capturing interaction through orthogonal projection, thus relieving modality imbalance phenomenon to some extent. However, the projection strategy orthogonal to the original space can lead to poor plasticity as the alternating learning proceeds, thus affecting model performance. To address this issue, in this paper, we propose a novel multimodal learning method called interactiveMML via flat gradient modification (IGM) by employing a flat gradient modification strategy to enhance interactive MML. Specifically, we first employ a flat projection-based gradient modification strategy that is independent to the original space, aiming to avoid the poor plasticity issue. Then we introduce the sharpness-aware minimization (SAM)-based optimization strategy to fully exploit the flatness of the learning objective and further enhance interaction during learning. To this end, the plasticity problem can be avoided and the overall performance is improved. Extensive experiments on widely used datasets demonstrate that IGM outperforms various state-of-the-art (SOTA) baselines, achieving superior performance. The source code is available at https://anonymous.4open.science/r/method-CC45.

The multimodal imbalance problem has been extensively studied to prevent the undesirable scenario where multimodal performance falls below that of unimodal models. However, existing methods typically assess the strength of modalities and perform learning simultaneously under the imbalanced status. This deferred strategy fails to rebalance multimodal learning instantaneously, leading to performance degeneration. To address this, we propose a novel multimodal learning approach, termed instantaneous probe-and-rebalance multimodal learning (IPRM), which employs a two-pass forward method to first probe (but not learn) and then perform rebalanced learning under the balanced status. Concretely, we first employ the geodesic multimodal mixup (GMM) to incorporate fusion representation and probe modality strength in the first forward phase. Then the weights are instantaneously recalibrated based on the probed strength, facilitating balanced training via the second forward pass. This process is applied dynamically throughout the entire training process. Extensive experiments reveal that our proposed IPRM outperforms all baselines, achieving state-of-the-art (SOTA) performance on numerous widely used datasets. The code is available at https://github.com/njustkmg/IJCAI25-IPRM.

Cross-modal retrieval refers to identifying semantically relevant data across different modalities. However, annotation errors or inherent ambiguity can cause semantic inconsistency in sample pairs, degrading retrieval performance. Prior efforts either relied heavily on the quality of explicitly dividing clean and noisy subsets, or solely leveraged carefully selected single anchor information, neglecting relationships among diverse neighbors. In this paper, we propose a novel Graph-based Label Propagation (GLP) framework that learns pseudo-labels via label propagation on a sparse graph, enabling self-correction of noisy labels. Specifically, each modality's instances are treated as nodes, connected via k-nearest neighbor (kNN) search to form a sparse graph. Pseudo-label vectors are generated for all nodes within one modality to capture the matching degree of inter-modal nodes. Through iterative label propagation, the stabilized pseudo-labels implicitly exploit both intra- and inter-modal relationships to derive a reliable matching degree. A dynamic queue further enhances graph quality by updating high-quality nodes. Experiments on Flickr30K, MSCOCO, and CC120K show that our method outperforms state-of-the-art approaches, especially under high noise. Code is available at https://github.com/njustkmg/MM25-GLP.

The integration of prompt tuning with multimodal learning has shown significant generalization abilities for various downstream tasks. Despite advancements, existing methods heavily depend on massive modality-specific labeled data (e.g., video, audio, and image), or are customized for a single modality. In this study, we present Text as Any-Modality by Consistent Prompt Tuning (TaAM-CPT), a scalable approach for constructing a general representation model toward unlimited modalities using solely text data. TaAM-CPT comprises modality prompt pools, text construction, and modality-aligned text encoders from pre-trained models, which allows for extending new modalities by simply adding prompt pools and modality-aligned text encoders. To harmonize the learning across different modalities, TaAM-CPT designs intra- and inter-modal learning objectives, which can capture category details within modalities while maintaining semantic consistency across different modalities. Benefiting from its scalable architecture and pre-trained models, TaAM-CPT can be seamlessly extended to accommodate unlimited modalities. Remarkably, without any modality-specific labeled data, TaAM-CPT achieves leading results on diverse datasets spanning various modalities, including video classification, image classification, and audio classification. The code is available at https://github.com/Jinx630/TaAM-CPT.

Multimodal learning (MML) is significantly constrained by modality imbalance, leading to suboptimal performance in practice. While existing approaches primarily focus on balancing the learning of different modalities to address this issue, they fundamentally overlook the inherent disproportion in model classification ability, which serves as the primary cause of this phenomenon. In this paper, we propose a novel multimodal learning approach to dynamically balance the classification ability of weak and strong modalities by incorporating the principle of boosting. Concretely, we first propose a sustained boosting algorithm in multimodal learning by simultaneously optimizing the classification and residual errors. Subsequently, we introduce an adaptive classifier assignment strategy to dynamically facilitate the classification performance of the weak modality. Furthermore, we theoretically analyze the convergence property of the cross-modal gap function, ensuring the effectiveness of the proposed boosting scheme. To this end, the classification ability of strong and weak modalities is expected to be balanced, thereby mitigating the imbalance issue. Empirical experiments on widely used datasets reveal the superiority of our method through comparison with various state-of-the-art (SOTA) multimodal learning baselines. The source code is available at https://github.com/njustkmg/NeurIPS25-AUG.

Multi-modal learning aims to enhance performance by unifying models from various modalities but often faces the "modality imbalance" problem in real data, leading to a bias towards dominant modalities and neglecting others, thereby limiting its overall effectiveness. To address this challenge, the core idea is to balance the optimization of each modality to achieve a joint optimum. Existing approaches often employ a modal-level control mechanism for adjusting the update of each modal parameter. However, such a global-wise updating mechanism ignores the different importance of each parameter. Inspired by subnetwork optimization, we explore a uniform sampling-based optimization strategy and find it more effective than global-wise updating. According to the findings, we further propose a novel importance sampling-based, element-wise joint optimization method, called Adaptively Mask Subnetworks Considering Modal Significance (AMSS). Specifically, we incorporate mutual information rates to determine the modal significance and employ non-uniform adaptive sampling to select foreground subnetworks from each modality for parameter updates, thereby rebalancing multi-modal learning. Additionally, we demonstrate the reliability of the AMSS strategy through convergence analysis. Building upon theoretical insights, we further enhance the multi-modal mask subnetwork strategy using unbiased estimation, referred to as AMSS+. Extensive experiments reveal the superiority of our approach over comparison methods.

Infrared (IR) images inherently face the dual challenges of noise contamination and reduced contrast. However, existing image enhancement methods often overlook the intrinsic correlations between these factors—noise and low contrast—during multistage enhancement processes. Consequently, this oversight leads to a significant reduction in the fidelity of intricate details in IR images. In this article, we present a synergistic IR image enhancement network that simultaneously achieves denoising, contrast improvement, and detail preservation (DCDNet), which breaks down the overall enhancement process into more manageable steps. DCDNet is comprised of a detail awareness unit (DAU), a deep denoising prior (DDP), and a contrast improvement module (CIM). To maintain the details in the IR image, DAU is developed to extract the original detail feature information in DDP and integrate them into the CIM during contrast improvement to improve the final result. The detail information is derived from the encoder of the DDP, which focuses on denoising. The preserved detail features are subsequently incorporated into the decoder of the CIM, which is dedicated to enhancing contrast. Experimental results validate that our proposed approach surpasses other state-of-the-art methods for enhancing IR images in terms of the peak signal-to-noise ratio (PSNR), structural similarity index (SSIM), visual information fidelity (VIF), and performance in downstream tasks. The code and dataset are publicly available athttps://github.com/ChickenEating/IR-Enhancement.

Remote sensing image captioning (RSIC) aims to generate concise and accurate textual descriptions for complex aerial scenes, yet faces a critical challenge of limited labeled training data. To address this data scarcity, semi-supervised RSIC that leverages unlabeled data (i.e., unlabeled images) has attracted significant attention. However, existing semi-supervised methods often rely on sample filtering strategies for handling unlabeled data, potentially discarding valuable information from unselected samples. To address this limitation, we propose U2DP, a novel semi-supervised remote sensing image captioning framework that comprehensively Unlocks Unlabeled Data Potential via adaptive supervision reconstruction. Specifically, we first leverage the pre-trained CLIP model to evaluate the expressive capability of the caption generator on unlabeled data, categorizing them into three levels, from fine-grained to coarse-grained: strongly matched, weakly matched, and unmatched. Via this evaluation, we construct different pseudo-supervision signals for unsupervised learning, including the model’s own predictions and the pseudo-label predicted by CLIP from a remote sensing vocabulary. Therefore, U2DP enables effective learning from unlabeled data without sample filtering, further improving model performance. Extensive experiments are conducted on three widely-used public RSIC benchmark datasets, including Sydney-Captions, UCM-Captions, and RSICD. The results consistently demonstrate that our U2DP is superior to existing methods across all datasets. The source code is publicly available at https://github.com/njustkmg/U2DP.

Job recommendation aims to provide potential talents with suitable job descriptions (JDs) consistent with their career trajectory, which plays an essential role in proactive talent recruitment. In real-world management scenarios, the available JD-user records always consist of JDs, user profiles, and click data, in which the user profiles are typically summarized as the user's skill distribution for privacy reasons. Although existing sophisticated recommendation methods can be directly employed, effective recommendation still has challenges considering the information deficit of JD itself and the natural heterogeneous gap between JD and user profile. To address these challenges, we proposed a novel skill-aware recommendation model based on the designed semantic-enhanced Transformer to parse JDs and complete personalized job recommendation. Specifically, we first model the relative items of each JD and then adopt an encoder with the local-global attention mechanism to better mine the intra-job and inter-job dependencies from JD tuples. Moreover, we adopt a two-stage learning strategy for skill-aware recommendation, in which we utilize the skill distribution to guide JD representation learning in the recall stage and then combine the user profiles for final prediction in the ranking stage. Consequently, we can embed rich contextual semantic representations for learning JDs, while skill-aware recommendation provides effective JD-user joint representation for click-through rate (CTR) prediction. To validate the superior performance of our method for job recommendation, we present a thorough empirical analysis of large-scale real-world and public datasets to demonstrate its effectiveness and interpretability.

Most existing biomedical language models are trained on plain text with general learning goals such as random word infilling, failing to capture the knowledge in the biomedical corpus sufficiently. Since biomedical articles usually contain many tables summarising the main entities and their relations, in the paper, we propose a Tabular knowledge enhanced bioMedical pretrained language model, called TabMedBERT. Specifically, we align entities between table cells, and article text spans with pre-defined rules. Then we add two table-related self-supervised tasks to integrate tabular knowledge into the language model: Entity Infilling (EI) and Table Cloze Test (TCT). While EI masks tokens within aligned entities in the article, TCT converts aligned entities in the table layout into a cloze text by erasing one entity and prompts the model to extract the appropriate span to fill in the blank. Experimental results demonstrate that TabMedBERT surpasses all competing language models without adding additional parameters, establishing a new state-of-the-art performance of 85.59% (+1.29%) on the BLURB biomedical NLP benchmark and 7 additional information extraction datasets. Moreover, the model architecture for TCT provides a straightforward solution to revise information extraction with paired entities.

Vision-language retrieval aims to perform cross-modal instances search by learning consistent vision-language representations. However, in real applications, vision-language divergence always results in strong and weak modalities and different modalities have various performances in uni-modal tasks. In this paper, we reveal that traditional vision-language hard consistency disrupts the relationships among uni-modal instances considering the weak-strong modal scenario, causing a decline in uni-modal retrieval capability. To address this issue, we propose Coordinated Vision-Language Retrieval (CoVLR), a solution that cooperatively optimizes both cross-modal consistency and intra-modal structure-preserving objectives via a meta-learning strategy. Specifically, CoVLR utilizes intra-modal structure-preserving as the meta-test task to validate the cross-modal consistency loss, which is considered the meta-train task. The effectiveness of CoVLR is validated through extensive experiments on commonly used datasets, which demonstrate superior results compared to other baselines on various retrieval scenarios.

Transferring knowledge across diverse data modalities is receiving increasing attention in machine learning. This paper tackles the task of leveraging expert-derived, yet expensive, tabular data to enhance image-based predictions when tabular data is unavailable during inference. The primary challenges stem from the inherent complexity of accurately mapping diverse tabular data to visual contexts, coupled with the necessity to devise distinct strategies for numerical and categorical tabular attributes. We propose CHannel tAbulaR alignment with optiMal tranSport (Charms), which establishes an alignment between image channels and tabular attributes, enabling selective knowledge transfer that is pertinent to visual features. Specifically, Charms measures similarity distributions across modalities to effectively differentiate and transfer relevant tabular features, with a focus on morphological characteristics, enhancing the capabilities of visual classifiers. By maximizing the mutual information between image channels and tabular features, knowledge from both numerical and categorical tabular attributes are extracted. Experimental results demonstrate that Charms not only enhances the performance of image classifiers but also improves their interpretability by effectively utilizing tabular knowledge.

Multimodal learning falls into the trap of the optimization dilemma due to the modality imbalance phenomenon, leading to unsatisfactory performance in real applications. A core reason for modality imbalance is that the models of each modality converge at different rates. Many attempts naturally focus on adjusting learning procedures adaptively. Essentially, the reason why models converge at different rates is because the difficulty of fitting category labels is inconsistent for each modality during learning. From the perspective of fitting labels, we find that appropriate positive intervention label fitting can correct this difference in learning ability. By exploiting the ability of contrastive learning to intervene in the learning of category label fitting, we propose a novel multimodal learning approach that dynamically integrates unsupervised contrastive learning and supervised multimodal learning to address the modality imbalance problem. We find that a simple yet heuristic integration strategy can significantly alleviate the modality imbalance phenomenon. Moreover, we design a learning-based integration strategy to integrate two losses dynamically, further improving the performance. Experiments on widely used datasets demonstrate the superiority of our method compared with state-of-the-art (SOTA) multimodal learning approaches. The code is available at https://github.com/njustkmg/NeurIPS24-LFM.

Open-set Semi-supervised Learning (OSSL) holds a realistic setting that unlabeled data may come from classes unseen in the labeled set, i.e., out-of-distribution (OOD) data, which could cause performance degradation in conventional SSL models. To handle this issue, except for the traditional in-distribution (ID) classifier, some existing OSSL approaches employ an extra OOD detection module to avoid the potential negative impact of the OOD data. Nevertheless, these approaches typically employ the entire set of open-set data during their training process, which may contain data unfriendly to the OSSL task that can negatively influence the model performance. This inspires us to develop a robust open-set data selection strategy for OSSL. Through a theoretical understanding from the perspective of learning theory, we propose Wise Open-set Semi-supervised Learning (WiseOpen), a generic OSSL framework that selectively leverages the open-set data for training the model. By applying a gradient-variance-based selection mechanism, WiseOpen exploits a friendly subset instead of the whole open-set dataset to enhance the model's capability of ID classification. Moreover, to reduce the computational expense, we also propose two practical variants of WiseOpen by adopting low-frequency update and loss-based selection respectively. Extensive experiments demonstrate the effectiveness of WiseOpen in comparison with the state-of-the-art.

The task of image captioning aims to generate captions directly from images via the automatically learned cross-modal generator. To build a well-performing generator, existing approaches usually need a large number of described images (i.e., supervised image-sentence pairs), requiring a huge effects on manual labeling. However, in real-world applications, a more general scenario is that we only have limited amount of described images and a large number of undescribed images. Therefore, a resulting challenge is how to effectively combine the undescribed images into the learning of cross-modal generator (i.e., semisupervised image captioning). To solve this problem, we propose a novel image captioning method by exploiting the cross-modal prediction and relation consistency (CPRC), which aims to utilize the raw image input to constrain the generated sentence in the semantic space. In detail, considering that the heterogeneous gap between modalities always leads to the supervision difficulty while using the global embedding directly, CPRC turns to transform both the raw image and corresponding generated sentence into the shared semantic space, and measure the generated sentence from two aspects: 1) prediction consistency: CPRC utilizes the prediction of raw image as soft label to distill useful supervision for the generated sentence, rather than employing the traditional pseudo labeling and 2) relation consistency: CPRC develops a novel relation consistency between augmented images and corresponding generated sentences to retain the important relational knowledge. In result, CPRC supervises the generated sentence from both the informativeness and representativeness perspectives, and can reasonably use the undescribed images to learn a more effective generator under the semisupervised scenario. The experiments show that our method outperforms state-of-the-art comparison methods on the MS-COCO "Karpathy" offline test split under complex nonparallel scenarios, for example, CPRC achieves at least 6% improvements on the CIDEr-D score.

Recent progress has shown that integrating multiscale visual features with advanced Transformer architectures is a promising approach for remote sensing image captioning (RSIC). However, the lack of local modeling ability in self-attention may potentially lead to inaccurate contextual information. Moreover, the scarcity of trainable image-caption pairs poses challenges in effectively harnessing the semantic alignment between images and texts. To mitigate these issues, we propose a Multiscale Grouping Transformer with Contrastive Language-Image Pre-training (CLIP) latents (MG-Transformer) for RSIC. First of all, a CLIP image embedding and a set of region features are extracted within a Multi-level Feature Extraction module. To achieve a comprehensive image representation, a Semantic Correlation module is designed to integrate the image embedding and region features with an attention gate. Subsequently, the integrated image features are fed into a Transformer model. The Transformer encoder utilizes dilated convolutions with different dilation rates to obtain multiscale visual features. To enhance the local modeling ability of the self-attention mechanism in the encoder, we introduce a Global Grouping Attention mechanism. This mechanism incorporates a grouping operation into self-attention, allowing each attention head to focus on different contextual information. The Transformer decoder then adopts the Meshed Cross-Attention mechanism to establish relationships between various scales of visual features and text features. This facilitates the generation of captions for images by the decoder. Experimental results on three RSIC datasets demonstrate the superiority of the proposed MG-Transformer. The code will be publicly available at https://github.com/One-paper-luck/MG-Transformer.

Vision-language retrieval aims to search for similar instances in one modality based on queries from another modality. The primary objective is to learn cross-modal matching representations in a latent common space. Actually, the assumption underlying cross-modal matching is modal balance, where each modality contains sufficient information to represent the others. However, noise interference and modality insufficiency often lead to modal imbalance, making it a common phenomenon in practice. The impact of imbalance on retrieval performance remains an open question. In this paper, we first demonstrate that ultimate cross-modal matching is generally sub-optimal for cross-modal retrieval when imbalanced modalities exist. The structure of instances in the common space is inherently influenced when facing imbalanced modalities, posing a challenge to cross-modal similarity measurement. To address this issue, we emphasize the importance of meaningful structure-preserved matching. Accordingly, we propose a simple yet effective method to rebalance cross-modal matching by learning structure-preserved matching representations. Specifically, we design a novel multi-granularity cross-modal matching that incorporates structure-aware distillation alongside the cross-modal matching loss. While the cross-modal matching loss constraints instance-level matching, the structure-aware distillation further regularizes the geometric consistency between learned matching representations and intra-modal representations through the developed relational matching. Extensive experiments on different datasets affirm the superior cross-modal retrieval performance of our approach, simultaneously enhancing single-modal retrieval capabilities compared to the baseline models.

In some real-world applications, data samples are usually distributed on local devices, where federated learning (FL) techniques are proposed to coordinate decentralized clients without directly sharing users’ private data. FL commonly follows the parameter server architecture and contains multiple personalization and aggregation procedures. The natural data heterogeneity across clients, i.e., Non-I.I.D. data, challenges both the aggregation and personalization goals in FL. In this paper, we focus on a special kind of Non-I.I.D. scene where clients own incomplete classes, i.e., each client can only access a partial set of the whole class set. The server aims to aggregate a complete classification model that could generalize to all classes, while the clients are inclined to improve the performance of distinguishing their observed classes. For better model aggregation, we point out that the standard softmax will encounter several problems caused by missing classes and propose “restricted softmax” as an alternative. For better model personalization, we point out that the hard-won personalized models are not well exploited and propose “inherited private model” to store the personalization experience. Our proposed algorithm named MAP could simultaneously achieve the aggregation and personalization goals in FL. Abundant experimental studies verify the superiorities of our algorithm.

Learning and development, or L&D, plays an important role in talent management, which aims to improve the knowledge and capabilities of employees through a variety of performance-oriented training activities. Recently, with the rapid development of enterprise management information systems, many research efforts and industrial practices have been devoted to building personalized employee training course recommender systems. Nevertheless, a widespread challenge is how to provide explainable recommendations with the consideration of different learning motivations from talents. To this end, we propose CKGE, a contextualized knowledge graph (KG) embedding approach for developing an explainable training course recommender system. A novel perspective of CKGE is to integrate both the contextualized neighbor semantics and high-order connections as motivation-aware information for learning effective representations of talents and courses. Specifically, in CKGE, for each entity pair (i.e., the talent-course pair), we first construct a meta-graph, including the neighbors of each entity and the meta-paths between entities as motivation-aware information. Then, we develop a novel KG-based Transformer, which can serialize entities and paths in the meta-graph as a sequential input, with the specially designed relational attention and structural encoding mechanisms to better model the global dependence of KG structured data. Meanwhile, the local path mask prediction can effectively reveal the importance of different paths. As a result, CKGE not only can make precise predictions but also can discriminate the saliencies of meta-paths in characterizing corresponding preferences. Extensive experiments on real-world and public datasets clearly validate the effectiveness and interpretability of CKGE compared with state-of-the-art baselines.

Existing semi-supervised learning (SSL) methods assume that labeled and unlabeled data share the same class space. However, in real-world applications, unlabeled data always contain classes not present in the labeled set, which may cause classification performance degradation of known classes. Therefore, open-world SSL approaches are researched to handle the presence of multiple unknown classes in the unlabeled data, which aims to accurately classify known classes while fine-grained distinguishing different unknown classes. To address this challenge, in this paper, we propose an open-world SSL method for Self-learning Open-world Classes (SSOC), which can explicitly self-learn multiple unknown classes. Specifically, SSOC first defines class center tokens for both known and unknown classes and autonomously learns token representations according to all samples with the cross-attention mechanism. To effectively discover novel classes, SSOC further designs a pairwise similarity loss in addition to the entropy loss, which can wisely exploit the information available in unlabeled data from instances’ predictions and relationships. Extensive experiments demonstrate that SSOC outperforms the state-of-the-art baselines on multiple popular classification benchmarks. Specifically, on the ImageNet-100 dataset with a novel ratio of $90 \%$, SSOC achieves a remarkable $16 \%$ improvement.

This paper presents a haptic device with a simple architecture of only two limbs that can provide translational motion in three degrees of freedom (DOF) and one-DOF rotational motion. Actuation redundancy eliminates all forward-kinematic singularities and improves the motion-force transmission property. Thanks to the special structure of the kinematic chains, all actuators are close to the base and full gravity compensation is achieved passively by using springs. Force producibility analysis shows that this haptic device is able to produce long-term continuous force feedback of 15–30 N in each direction. By developing a prototype of the haptic device and a virtual three-dimensional simulator, a preliminary performance evaluation of the haptic device was conducted. In addition, a torque distribution algorithm considering a relaxed form of actuator-torque saturation was experimentally evaluated, and a comparison with other algorithms reveals that this algorithm offers several advantages.

Aiding humans with scientific designs is one of the most exciting of artificial intelligence (AI) and machine learning (ML), due to their potential for the discovery of new drugs, design of new materials and chemical compounds, etc. However, scientific design typically requires complex domain knowledge that is not familiar to AI researchers. Further, scientific studies involve professional skills to perform experiments and evaluations. These obstacles prevent AI researchers from developing specialized methods for scientific designs. To take a step towards easy-to-understand and reproducible research of scientific design, we propose a benchmark for the inverse design of nanophotonic devices, which can be verified computationally and accurately. Specifically, we implemented three different nanophotonic design problems, namely a radiative cooler, a selective emitter for thermophotovoltaics, and structural color filters, all of which are different in design parameter spaces, complexity, and design targets. The benchmark environments are implemented with an open-source simulator. We further implemented 10 different inverse design algorithms and compared them in a reproducible and fair framework. The results revealed the strengths and weaknesses of existing methods, which shed light on several future directions for developing more efficient inverse design algorithms. Our benchmark can also serve as the starting point for more challenging scientific design problems. The code of IDToolkit is available at https://github.com/ThyrixYang/IDToolkit.

The timely prediction of bankruptcy is highly desirable to guarantee an upward spiral for overall societal well-being. By extracting multifaceted information from the business interaction networks, Graph Neural Networks (GNNs) may be able to automatically make more informed predictions for bankruptcy, as compared to methods that rely heavily on abundant manpower to a large extent. Yet in real applications, bankruptcy prediction faces the key issue of quantity-imbalance: data usually comes with a long-tailed distribution wherein bankrupt corporates occupy the least of the data proportion but are our target to be identified. Apart from that, the topology-imbalance issue behind graph-structural data exacerbates prediction deterioration: feature propagation is dominated by non-bankrupt nodes through messages passing between nodes; thus, bankrupt nodes receive highly confusing information and could be easily assimilated by nearby non-bankrupt nodes. Unfortunately, the existing GNN methods are not immune to these two imbalance issues. To tackle the challenging but practically useful scenario, we propose a novel bankruptcy prediction model called the Quantity and Topology Imbalance-Aware Heterogeneous Graph Neural Network (QTIAH-GNN) to boost the final performance. Specifically, QTIAH-GNN employs the multi-hierarchy label-aware neighbor selection to conquer the topology-imbalance issue by using the class-semantic representation and the learnable parameterized similarity metric, and employs the imbalance-oriented loss to obtain the optimal tradeoff between the accuracies of the majority and minority classes. In experiments, we evaluate the proposed QTIAH-GNN on two large-scale, real-world datasets. The results show that QTIAH-GNN outperforms other state-of-the-art baselines in terms of prediction accuracy with superior efficiency and generalization ability, has stronger robustness to data imbalance, and provides meaningful model interpretation.

Active learning (AL) methods have been proven to be an effective way to reduce the labeling effort by intelligently selecting valuable instances for annotation. Despite their great success with in-distribution (ID) scenarios, AL methods suffer from performance degradation in many real-world applications because out-of-distribution (OOD) instances are always inevitably contained in unlabeled data, which may lead to inefficient sampling. Therefore, several attempts have been explored open-set AL by strategically selecting pure ID instances while filtering OOD instances. However, concentrating solely on selecting pseudo-ID instances may cause the training constraint of the ID classifier and OOD detector. To address this issue, we propose a simple yet effective sampling scheme, Progressive Active Learning (PAL), which employs a progressive sampling mechanism to leverage the active selection of valuable OOD instances. The proposed PAL measures unlabeled instances by synergistically evaluating instances' informativeness and representativeness, and thus it can balance the pseudo-ID and pseudo-OOD instances in each round to enhance both the capacity of the ID classifier and the OOD detector. %Meanwhile, PAL measures unlabeled instances by synergistically evaluating instances' informativeness and representativeness, which can more effectively estimate the values of instances. Extensive experiments on various open-set AL scenarios demonstrate the effectiveness of the proposed PAL, compared with the state-of-the-art methods. The code is available at \url{https://github.com/njustkmg/PAL}.

Remote sensing image captioning aims to generate meaningful and grammatically accurate sentences for remote sensing images. However, in comparison to natural image captioning, remote sensing image captioning encounters additional challenges due to the unique characteristics of remote sensing images. The first challenge arises from the abundance of objects present in these images. As the number of objects increases, it becomes increasingly difficult to determine the main focus of the description. Moreover, the objects in remote sensing images often share similar appearances, which further complicates the generation of accurate descriptions. To overcome these challenges, we propose a Prior Knowledge-guided Transformer for remote sensing image captioning. Firstly, scene-level and object-level features are extracted in a Multi-level Feature Extraction module. To further refine and enhance the extracted multi-level features, we introduce a Feature Enhancement module. This module utilizes a combination of graph neural networks and attention mechanisms to capture the correlation and difference between different objects or scene regions. Moreover, we propose a Prior Knowledge augmented Attention mechanism to select the objects that are more relevant to the scene regions by establishing the relationships between them. This attention mechanism is seamlessly integrated into the Transformer structure, providing valuable prior knowledge that promotes the caption generation process. Extensive experiments on three remote sensing image captioning datasets verify the superiority of the proposed method. Compared with the baseline methods, the proposed method achieves more impressive performance. The code will be publicly available at https://github.com/One-paper-luck/PKG-Transformer.

Class-Incremental Learning (CIL) aims to train a reliable model with the streaming data, which emerges unknown classes sequentially. Different from traditional closed set learning, CIL has two main challenges: (1) Novel class detection. The initial training data only contains incomplete classes, and streaming test data will accept unknown classes. Therefore, the model needs to not only accurately classify known classes, but also effectively detect unknown classes; (2) Model expansion. After the novel classes are detected, the model needs to be updated without re-training using the entire previous data. However, traditional CIL methods have not fully considered these two challenges. To this end, we propose a Class-Incremental Learning without Forgetting (CILF) framework. In detail, CILF designs to regularize classification with decoupled prototype based loss, which can improve the intra-class and inter-class structure significantly, and acquire a compact embedding representation for novel class detection in result. Then, CILF employs a learnable curriculum clustering operator to estimate the number of semantic clusters via fine-tuning the learned network, in which curriculum operator can adaptively learn the embedding in self-taught form. Last, with the labeled streaming test data, CILF can update the network with robust regularization to mitigate the catastrophic forgetting.

Corporate relative valuation (CRV) refers to the process of comparing a company's value from company products, core staff and other related information, so that we can assess the company's market value, which is critical for venture capital firms. Traditionally, relative valuation methods heavily rely on tedious and expensive human efforts, especially for non-publicly listed companies. However, the availability of information about company's invisible assets, such as patents, talent, and investors, enables a new paradigm for learning and evaluating corporate relative values automatically. Indeed, in this paper, we reveal that, if the companies and their core members are formed as a heterogeneous graph and the attributes of different nodes include semantically-rich multi-modal data, it is able to extract a latent embedding for each company. Along this line, we develop an end-to-end heterogeneous multi-modal graph neural network method, named HM$^2$. Specifically, HM$^2$firstly perform the representation learning for heterogeneous neighbors of input company by taking relationships among nodes into consideration, which aggregates node attributes via linkage-aware multi-head attention mechanism, rather than multi-instance based methods. Then, HM$^2$adopts the self-attention network to aggregate different modal embeddings for final prediction, and employs dynamic triplet loss with embeddings of competitors as the constraint.

Most existing approaches often utilize the pre-fixed structure and large number of labeled data for training complex deep models, which are difficult to implement on incremental scenarios. As a matter of fact, real-world data is always in stream form. Thereby, there exits two challenges for building incremental deep models: a)Capacity Scalability. The entire training data is not available before learning the task. It is a challenge to make the deep model structure scale with streaming data for flexible model evolution and faster convergence. b)Capacity Sustainability. The distribution streaming data usually changes in nature (concept drift), thus it is necessary to update the model while preserving previous knowledge for overcoming the catastrophic forgetting. To this end, we develop an incremental deep model (IDM), which expands the network structure according to streaming data and slows down forgetting with the adaptive fisher regularization. However, IDM ignores another significant challenge with streaming data: c)Capacity Demand. Training a deep model always needs a large amount of labeled data, whereas it is almost impossible to label all unlabeled instances in real time. The core problem is to select a small number of the most discriminative instances to label while keeping the predictive accuracy of the model. Thereby, we focus on the online semi-supervised learning scenario with abrupt changes in data distribution, and further improve IDM to a cost-effective incremental deep model (CE-IDM), which can adaptively select the most discriminative newly coming instances for query to reduce the manual labeling costs. Specifically, CE-IDM adopts a novel extensible deep network structure by using an extra attention model for hidden layers. Based on the adaptive attention weights, CE-IDM develops a novel instance selection criterion by jointly estimating unlabeled instances’ representative and informative degree to satisfy the capacity demand. With the newly labeled instances, CE-IDM can quickly update the model with adaptive depth from streaming data and enable capacity scalability. Also, we address capacity sustainability by exploiting the attention based fisher information matrix, which can slow down the forgetting in consequence. Finally, CE-IDM can deal with the three capacity challenges methioned above in a unified framework. We conduct extensive experiments on real-world data and show that CE-IDM outperforms the state-of-the-art methods with a substantial margin.

Deep semi-supervised learning (SSL) aims to utilize a sizeable unlabeled set to train deep networks, thereby reducing the dependence on labeled instances. However, the unlabeled set often carries unseen classes that cause the deep SSL algorithm to lose generalization. Previous works focus on the data level that they attempt to remove unseen class data or assign lower weight to them but could not eliminate their adverse effects on the SSL algorithm. Rather than focusing on the data level, this paper turns attention to the model parameter level. We find that only partial parameters are essential for seen-class classification, termed safe parameters. In contrast, the other parameters tend to fit irrelevant data, termed harmful parameters. Driven by this insight, we propose Safe Parameter Learning (SPL) to discover safe parameters and make the harmful parameters inactive, such that we can mitigate the adverse effects caused by unseen-class data. Specifically, we firstly design an effective strategy to divide all parameters in the pre-trained SSL model into safe and harmful ones. Then, we introduce a bi-level optimization strategy to update the safe parameters and kill the harmful parameters. Extensive experiments show that SPL outperforms the state-of-the-art SSL methods on all the benchmarks by a large margin. Moreover, experiments demonstrate that SPL can be integrated into the most popular deep SSL networks and be easily extended to handle other cases of class distribution mismatch.

In talent management, resume assessment aims to analyze the quality of a job seeker's resume, which can assist recruiters to discover suitable candidates and benefit job seekers improving resume quality in return. Recent machine learning based methods on large-scale public resume datasets have provided the opportunity for automatic assessment for reducing manual costs. However, most existing approaches are still content-dominated and ignore other valuable information. Inspired by practical resume evaluations that consider both the content and layout, we construct the multi-modalities from resumes but face a new challenge that sometimes the performance of multi-modal fusion is even worse than the best uni-modality. In this paper, we experimentally find that this phenomenon is due to the cross-modal divergence. Therefore, we need to consider when is it appropriate to perform multi-modal fusion? To address this problem, we design an instance-aware fusion method, i.e., Divergence-Orientated Multi-Modal Fusion Network (DOMFN), which can adaptively fuse the uni-modal predictions and multi-modal prediction based on cross-modal divergence. Specifically, DOMFN computes a functional penalty score to measure the divergence of cross-modal predictions. Then, the learned divergence can be used to decide whether to conduct multi-modal fusion and be adopted into an amended loss for reliable training. Consequently, DOMFN rejects multi-modal prediction when the cross-modal divergence is too large, avoiding the overall performance degradation, so as to achieve better performance than uni-modalities. In experiments, qualitative comparison with baselines on real-world dataset demonstrates the superiority and explainability of the proposed DOMFN, e.g., we find a meaningful phenomenon that multi-modal fusion has positive effects for assessing resumes from UI Designer and Enterprise Service positions, whereas affects the assessment of Technology and Product Operation positions.

Open set classification (OSC) tackles the problem of determining whether the data are in-class or out-of-class during inference, when only provided with a set of in-class examples at training time. Traditional OSC methods usually train discriminative or generative models with the owned in-class data, and then utilize the pre-trained models to classify test data directly. However, these methods always suffer from the embedding confusion problem, i.e., partial out-of-class instances are mixed with in-class ones of similar semantics, making it difficult to classify. To solve this problem, we unify semi-supervised learning to develop a novel OSC algorithm, S2OSC, which incorporates out-of-class instances filtering and model re-training in a transductive manner. In detail, given a pool of newly coming test data, S2OSC firstly filters the mostly distinct out-of-class instances using the pre-trained model, and annotates super-class for them. Then, S2OSC trains a holistic classification model by combing in-class and out-of-class labeled data with the remaining unlabeled test data in a semi-supervised paradigm. Furthermore, considering that data are usually in the streaming form in real applications, we extend S2OSC into an incremental update framework (I-S2OSC), and adopt a knowledge memory regularization to mitigate the catastrophic forgetting problem in incremental update. Despite the simplicity of proposed models, the experimental results show that S2OSC achieves state-of-the-art performance across a variety of OSC tasks, including 85.4% of F1 on CIFAR-10 with only 300 pseudo-labels. We also demonstrate how S2OSC can be expanded to incremental OSC setting effectively with streaming data.

New class detection and effective model expansion are of great importance in incremental data mining. In open incremental data environments, data often come with novel classes, e.g., the emergence of new classes in image classification or new topics in opinion monitoring, and is denoted as class-incremental learning (C-IL) in literature. There are two main challenges in C-IL: how to conduct novelty detection and how to update the model with few novel class instances. Most previous methods pay much attention to the former challenge while ignoring the problem of efficiently updating models. To solve this problem, we propose a novel framework to handle the incremental new class, named learning to classify with incremental new class (LC-INC), which can process these two challenges automatically in one unified framework. In detail, LC-INC utilizes a novel structure network to consider the prototype information between class centers of known classes and newly incoming instances, which can dynamically combine the prediction information with structure information to detect novel class instances efficiently. On the other hand, the proposed structure network can also act as a meta-network, which can learn to expand the model much faster and more efficiently with inadequate novel class instances. Experiments on synthetic and real-world datasets successfully validate the effectiveness of our proposed method.

Estimated time of arrival (ETA) is a very important factor in the transportation system. It has attracted increasing attentions and has been widely used as a basic service in navigation systems and intelligent transportation systems. In this paper, we propose a novel solution to the ETA estimation problem, which is an ensemble on tree models and neural networks. We proved the accuracy and robustness of the solution on the A/B list and finally won first place in the SIGSPATIAL 2021 GISCUP competition.

The main challenge of cross-modal retrieval is to learn the consistent embedding for heterogeneous modalities. To solve this problem, traditional label-wise cross-modal approaches usually constrain the inter-modal and intra-modal embedding consistency relying on the label ground-truths. However, the experiments reveal that different modal networks actually have various generalization capacities, thereby end-to-end joint training with consistency loss usually leads to sub-optimal uni-modal model, which in turn affects the learning of consistent embedding. Therefore, in this paper, we argue that what really needed for supervised cross-modal retrieval is a good shared classification model. In other words, we learn the consistent embedding by ensuring the classification performance of each modality on the shared model, without the consistency loss. Specifically, we consider a technique called Semantic Sharing, which directly trains the two modalities interactively by adopting a shared self-attention based classification model. We evaluate the proposed approach on three representative datasets. The results validate that the proposed semantic sharing can consistently boost the performance under NDCG metric.

This paper presents an underactuated robotic gripper that consists of three fingers. This gripper is driven by seven actuators and capable of grasping a wide range of objects in different working scenarios. A combination of a four-bar mechanism and parallelograms ensures that each finger can provide the basic pinch grasp and power grasp. Detailed fingertip grasping force analysis shows the large payload of this gripper. To fulfill multiple challenging grasping tasks, the fingertip orientation of each finger was designed to be decoupled from the finger flexion motion. Particular emphasis is placed on the environmental contact-based grasp and active transition from the pinch grasp to the power grasp. Detailed analysis shows that the contact-based fingertip grasp could be used to grasp thin objects lying on a flat surface with widths less than 168 mm safely and stably. To prevent over-squeezing of the grasped object in the active transition, the fingertip orientation and finger flexion motion were controlled coordinately. Moreover, a combination of the contact-based grasp and the direct power grasp was also allowed to grasp objects lying on a flat surface and having smooth surfaces robustly. Experimental results demonstrate the effectiveness of the proposed gripper in real-world applications.

Query and Point-of-Interest (POI) matching, aiming at recommending the most relevant POIs from partial query keywords, has become one of the most essential functions in online navigation and ride-hailing applications. Existing methods for query-POI matching, such as Google Maps and Uber, have a natural focus on measuring the static semantic similarity between contextual information of queries and geographical information of POIs. However, it remains challenging for dynamic and personalized online query-POI matching because of the non-stationary and situational context-dependent query-POI relevance. Moreover, the large volume of online queries requires an adaptive and incremental model training strategy that is efficient and scalable in the online scenario. To this end, in this paper, we propose an Incremental Spatio-Temporal Graph Learning (IncreSTGL) framework for intelligent online query-POI matching. Specifically, we first model dynamic query-POI interactions as microscopic and macroscopic graphs. Then, we propose an incremental graph representation learning module to refine and update query-POI interaction graphs in an online incremental fashion, which includes: (i) a contextual graph attention operation quantifying query-POI correlation based on historical queries under dynamic situational context, (ii) a graph discrimination operation capturing the sequential query-POI relevance drift from a holistic view of personalized preference and social homophily, and (iii) a multi-level temporal attention operation summarizing the temporal variations of query-POI interaction graphs for subsequent query-POI matching. Finally, we introduce a lightweight semantic matching module for online query-POI similarity measurement. To demonstrate the effectiveness and efficiency of the proposed algorithm, we conduct extensive experiments on two real-world datasets collected from a leading online navigation and map service provider in China.

Complex objects are usually with multiple labels, and can be represented by multiple modal representations, e.g., the complex articles contain text and image information as well as multiple annotations. Previous methods assume that the homogeneous multi-modal data are consistent, while in real applications, the raw data are disordered, e.g., the article constitutes with variable number of inconsistent text and image instances. Therefore, Multi-modal Multi-instance Multi-label (M3) learning provides a framework for handling such task and has exhibited excellent performance. However, M3 learning is facing two main challenges: 1) how to effectively utilize label correlation and 2) how to take advantage of multi-modal learning to process unlabeled instances. To solve these problems, we first propose a novel Multi-modal Multi-instance Multi-label Deep Network (M3DN), which considers M3 learning in an end-to-end multi-modal deep network and utilizes consistency principle among different modal bag-level predictions. Based on the M3DN, we learn the latent ground label metric with the optimal transport. Moreover, we introduce the extrinsic unlabeled multi-modal multi-instance data, and propose the M3DNS, which considers the instance-level auto-encoder for single modality and modified bag-level optimal transport to strengthen the consistency among modalities. Thereby M3DNS can better predict label and exploit label correlation simultaneously. Experiments on benchmark datasets and real world WKG Game-Hub dataset validate the effectiveness of the proposed methods.

In this paper, we propose a novel Semi-supervised Learning with Incomplete Modality (SLIM) method considering the modal consistency and complementarity simultaneously, and Kernel SLIM (SLIM-K) based on matrix completion for further solving the modal incompleteness. As is well known, most realistic data have multi-modal representations, multi-modal learning refers to the process of learning a precise model for complete modalities. However, due to the failures of data collection, self-deficiencies, or other various reasons, multi-modal examples are usually with incomplete modalities, which generate utility obstacle using previous methods. In this paper, SLIM integrates the intrinsic consistency and extrinsic complementary information for prediction and cluster simultaneously. In detail, SLIM forms different modal classifiers and clustering learner consistently in a unified framework, while using the extrinsic complementary information from unlabeled data against the insufficiencies brought by the incomplete modal issue. Moreover, in order to deal with missing modality in essence, we propose the SLIM-K, which takes the complemented kernel matrix into the classifiers and the cluster learner respectively. Thus, SLIM-K can solve the defects of missing modality in result. Finally, we give the discussion of generalization of incomplete modalities. Experiments on 13 benchmark multi-modal datasets and two real-world incomplete multi-modal datasets validate the effectiveness of our methods.

The next-item recommendation has attracted great research interests with both static and dynamic users' preferences considered. Existing approaches typically utilize user-item binary relations, and assume a flat preference distribution over items for each user. However, this assumption neglects the hierarchical discrimination between user intentions and user preferences, causing the methods have limited capacity to depict intention-specific preference. In fact, a consumer's purchasing behavior involves a natural sequential process, i.e., he/she first has an intention to buy one type of items, followed by choosing a specific item according to his/her preference under this intention. To this end, we propose a novel key-array memory network (KA-MemNN), which takes both user intentions and preferences into account for next-item recommendation. Specifically, the user behavioral intention tendency is determined through key addressing. Further, each array outputs an intention-specific preference representation of a user. Then, the degree of user's behavioral intention tendency and intention-specific preference representation are combined to form a hierarchical representation of a user. This representation is further utilized to replace the static profile of users in traditional matrix factorization for the purposes of reasoning. The experimental results on real-world data demonstrate the advantages of our approach over state-of-the-art methods.

In real-world applications, data are often with multiple modalities, and many multi-modal learning approaches are proposed for integrating the information from different sources. Most of the previous multi-modal methods utilize the modal consistency to reduce the complexity of the learning problem, therefore the modal completeness needs to be guaranteed. However, due to the data collection failures, self-deficiencies, and other various reasons, multi-modal instances are often incomplete in real applications, and have the inconsistent anomalies even in the complete instances, which jointly result in the inconsistent problem. These degenerate the multi-modal feature learning performance, and will finally affect the generalization abilities in different tasks. In this paper, we propose a novel Deep Robust Unsupervised Multi-modal Network structure (DRUMN) for solving this real problem within a unified framework. The proposed DRUMN can utilize the extrinsic heterogeneous information from unlabeled data against the insufficiency caused by the incompleteness. On the other hand, the inconsistent anomaly issue is solved with an adaptive weighted estimation, rather than adjusting the complex thresholds. As DRUMN can extract the discriminative feature representations for each modality, experiments on real-world multimodal datasets successfully validate the effectiveness of our proposed method.

Since the financial crisis in late 2008-2009, several global regulatory authorities have mandated stress-testing exercises to evaluate the potential capital shortfalls & systemic impacts that large banks may face during adverse economic conditions. Thus, having the ability to analyze economic conditions & banking performance profiles together to determine relationships among their respective features may provide insights for stress testing tasks. In this paper, we propose an Integrated Multimodal Bank Stress Test Prediction (IMBSTP) model framework consisting of a two-stages; (1) economic conditions estimator to approximate joint representation among the exogenous factors using generative models, (2) bank capital & loss forecaster to project stress test measures based on dimensional & temporal features selected from the exogenous economic conditions & banking performance profiles using a dual-attention recurrent neural network. Extensive experimentation is performed on historical economic conditions & consolidated financial statements of U.S. bank holdings companies to show the effectiveness of our approach when compared to state-of-the-art baseline methods.

Multi-modal learning refers to the process of learning a precise model to represent the joint representations of different modalities. Despite its promise for multi-modal learning, the co-regularization method is based on the consistency principle with a sufficient assumption, which usually does not hold for real-world multi-modal data. Indeed, due to the modal insufficiency in real-world applications, there are divergences among heterogeneous modalities. This imposes a critical challenge for multi-modal learning. To this end, in this paper, we propose a novel Comprehensive Multi-Modal Learning (CMML) framework, which can strike a balance between the consistency and divergency modalities by considering the insufficiency in one unified framework. Specifically, we utilize an instance level attention mechanism to weight the sufficiency for each instance on different modalities. Moreover, novel diversity regularization and robust consistency metrics are designed for discovering insufficient modalities. Our empirical studies show the superior performances of CMML on real-world data in terms of various criteria.

Recent years have witnessed growing interests in developing deep models for incremental learning. However, existing approaches often utilize the fixed structure and online backpropagation for deep model optimization, which is difficult to be implemented for incremental data scenarios. Indeed, for streaming data, there are two main challenges for building deep incremental models. First, there is a requirement to develop deep incremental models with Capacity Scalability. In other words, the entire training data are not available before learning the task. It is a challenge to make the deep model structure scaling with streaming data for flexible model evolution and faster convergence. Second, since the stream data distribution usually changes in nature (concept drift), there is a constraint for Capacity Sustainability. That is, how to update the model while preserving previous knowledge for overcoming the catastrophic forgetting. To this end, in this paper, we develop an incremental adaptive deep model (IADM) for dealing with the above two capacity challenges in real-world incremental data scenarios. Specifically, IADM provides an extra attention model for the hidden layers, which aims to learn deep models with adaptive depth from streaming data and enables capacity scalability. Also, we address capacity sustainability by exploiting the attention based fisher information matrix, which can prevent the forgetting in consequence. Finally, we conduct extensive experiments on real-world data and show that IADM outperforms the state-of-the-art methods with a substantial margin. Moreover, we show that IADM has better capacity scalability and sustainability in incremental learning scenarios.

Software Quality Assurance (SQA) is essential in software development and many defect prediction methods based on machine learning have been proposed to identify defective modules. However, most existing defect prediction models do not provide good defect prediction results, and the semantic features reflecting the detective patterns may not be well-captured via traditional feature extraction methods. More information such as code comments should be also be embedded to generate semantic features respecting the source code functionality. Therefore, how to embed code comments for defect prediction is a big challenge, and another problem is that many comments of source code are missing in real-world applications. In this paper, we propose a novel defect prediction model named CAP-CNN (Convolutional Neural Network for Comments Augmented Programs), which is a deep learning model that automatically embeds code comments in generating semantic features from the source code for software defect prediction. To overcome the missing comments problem, a novel training strategy is used in CAP-CNN that the network encodes and absorb comments information to generate semantic features automatically during training process, which does not need testing modules to contain comments. Experimental results on several widely-used software data sets indicate that the comment features are able to improve defect prediction performance.

In real world applications, data are often with multiple modalities. Researchers proposed the multi-modal learning approaches for integrating the information from different modalities. Most of the previous multi-modal methods assume that training examples are with complete modalities. However, due to the failures of data collection, self-deficiencies and other various reasons, multi-modal examples are usually with incomplete feature representation in real applications. In this paper, the incomplete feature representation issues in multi-modal learning are named as incomplete modalities, and we propose a semi-supervised multi-modal learning method aimed at this incomplete modal issue (SLIM). SLIM can utilize the extrinsic information from unlabeled data against the insufficiencies brought by the incomplete modal issues in a semi-supervised scenario. Besides, the proposed SLIM forms the problem into a unified framework which can be treated as a classifier or clustering learner, and integrate the intrinsic consistencies and extrinsic unlabeled information. As SLIM can extract the most discriminative predictors for each modality, experiments on 15 real world multi-modal datasets validate the effectiveness of our method.

In real world applications, complex objects are usually with multiple labels, and can be represented as multiple modal representations, e.g., the complex articles contain text and image information as well as are with multiple annotations. Previous methods assume that the homogeneous multi-modal data are consistent, while in real applications, the raw data are disordered, i.e., the article is constituted with variable number of inconsistent text and image instances. To solve this problem, Multi-modal Multi-instance Multi-label (M3) learning provides a framework for handling such task and has exhibited excellent performance. Besides, how to effectively utilize label correlation is also a challenging issue. In this paper, we propose a novel Multi-modal Multi-instance Multi-label Deep Network (M3DN), which learns the label prediction and exploits label correlation simultaneously based on the Optimal Transport, by considering the consistency principle between different modal bag-level prediction and the learned latent ground label metric. Experiments on benchmark datasets and real world WKG Game-Hub dataset validate the effectiveness of the proposed method.

Model reuse attempts to construct a model by utilizing existing available models, mostly trained for other tasks, rather than building a model from scratch. It is helpful to reduce the time cost, data amount, and expertise required. Deep learning has achieved great success in various tasks involving images, voices and videos. There are several studies have the sense of model reuse, by trying to reuse pre-trained deep networks architectures or deep model features to train a new deep model. They, however, neglect the fact that there are many other fixed models or features available. In this paper, we propose a more thorough model reuse scheme, FMR (Fixed Model Reuse). FMR utilizes the learning power of deep models to implicitly grab the useful discriminative information from fixed model/features that have been widely used in general tasks. We firstly arrange the convolution layers of a deep network and the provided fixed model/features in parallel, fully connecting to the output layer nodes. Then, the dependencies between the output layer nodes and the fixed model/features are knockdown such that only the raw feature inputs are needed when the model is being used for testing, though the helpful information in the fixed model/features have already been incorporated into the model. On one hand, by the FMR scheme, the required amount of training data can be significantly reduced because of the reuse of fixed model/features. On the other hand, the fixed model/features are not explicitly used in testing, and thus, the scheme can be quite useful in applications where the fixed model/features are protected by patents or commercial secrets. Experiments on five real-world datasets validate the effectiveness of FMR compared with state-of-the-art deep methods.

With the fast development of data collection techniques, a huge amount of complex multi-modal data are generated, shared and stored on the Internet. The burden of extracting multi-modal features for test instances in data analysis becomes the main fact that hurts the efficiency of prediction. In this paper, in order to reduce the modal extraction cost in serialized classification system, we propose a novel end-to-end serialized adaptive decision approach named Discriminative Modal Pursuit (\sc Dmp), which can automatically extract instance-specifically discriminative modal sequence for reducing the cost of feature extraction in the test phase. Rather than jointly optimize a highly non-convex empirical risk minimization problem, we are inspired by LSTM, and the proposed \sc Dmp can turn to learn the decision policies which predict the label information and decide the modalities to be extracted simultaneously within limited modal acquisition budget. Consequently, \sc Dmp approach can balance the classification performance and modal feature extraction cost by utilizing different modalities for different test instances. Empirical studies show that \sc Dmp is more efficient and effective than existing modal/feature extraction methods.

Multi-Model Reuse is one of the prominent problems in Learnware framework, while the main issue of Multi-Model Reuse lies in the final prediction acquisition from the responses of multiple pre-trained models. Different from multi-classifiers ensemble, there are only pre-trained models rather than the whole training sets provided in Multi-Model Reuse configuration. This configuration is closer to the real applications where the reliability of each model cannot be evaluated properly. In this paper, aiming at the lack of evaluation on reliability, the potential consistency spread on different modalities is utilized. With the consistency of pre-trained models on different modalities, we propose a Pre-trained Multi-Model Reuse approach PM2R with multi-modal data, which realizes the reusability of multiple models. PM2R can combine pre-trained multi-models efficiently without re-training, and consequently no more training data storage is required. We describe the more realistic Multi-Model Reuse setting comprehensively in our paper, and point out the differences among this setting, classifier ensemble and later fusion on multi-modal learning. Experiments on synthetic and real-world datasets validate the effectiveness of PM2R when it is compared with state-of-the-art ensemble/multi-modal learning methods under this more realistic setting.