This paper presents a novel approach for generating high-quality, cross-category 3D models from free-hand sketches with limited training data. We propose the first semi-supervised learning method to our knowledge for sketch-to-3D model conversion. Innovatively, we design a coarse-to-fine pipeline to perform the semi-supervised learning in the coarse stage and train a diffusion-based refiner to get a high-resolution 3D model. We designed a sketch-augmentation method for semi-supervised learning and integrated priors such as CLIP loss, shape prototypes, and adversarial loss to help generate high-quality results even with abstract and imprecise sketches. We also introduce an innovative procedural 3D generation method based on CAD code, which helps pre-train part of the network before fine-tuning with limited real data. Our approach, coupled with a specifically designed curriculum learning, allows us to generate high-quality 3D models across multiple categories with as few as 300 sketch-3D model pairs, marking a significant advancement over previous single-category approaches. In addition, we introduce the KO2D dataset, the largest collection of hand-drawn sketch-3D pairs to support further research in this area. As sketches are a far more intuitive and detailed way for users to express their unique ideas, we believe that this paper can move us closer to democratizing 3D content creation, enabling anyone to transform their ideas into 3D models effortlessly.

Large language models (LLMs) achieve impressive results on advanced mathematics benchmarks but sometimes fail on basic arithmetic tasks, raising the question of whether they have truly grasped fundamental arithmetic rules or are merely relying on pattern matching.To unravel this issue, we systematically probe LLMs' understanding of two-integer addition (0 to 2 64 ) by testing three crucial properties: commutativity (A + B = B + A), representation invariance via symbolic remapping (e.g., 7 → Y), and consistent accuracy scaling with operand length.Our evaluation of 12 leading LLMs reveals a stark disconnect: while models achieve high numeric accuracy (73.8-99.8%),they systematically fail these diagnostics.Specifically, accuracy plummets to ≤ 7.5% with symbolic inputs, commutativity is violated in up to 20% of cases, and accuracy scaling is non-monotonic.Interventions further expose this pattern-matching reliance: explicitly providing rules degrades performance by 29.49%, while prompting for explanations before answering merely maintains baseline accuracy.These findings demonstrate that current LLMs address elementary addition via pattern matching, not robust rule induction, motivating new diagnostic benchmarks and innovations in model architecture and training to cultivate genuine mathematical reasoning.we release both our diagnostic dataset and the code for dataset generation at https://github.com/ kuri-leo/llm-arithmetic-diagnostic.

There is a growing interest in applying machine learning techniques to healthcare. Recently, federated machine learning (FL) is gaining popularity since it allows researchers to train powerful models without compromising data privacy and security. However, the performance of existing FL approaches often deteriorates when encountering non-iid situations where there exist distribution gaps among clients, and few previous efforts focus on personalization in healthcare. In this article, we propose FedAP to tackle domain shifts and obtain personalized models for local clients. FedAP learns the similarity between clients via the statistics of the batch normalization layers while preserving the specificity of each client with different local batch normalization. Comprehensive experiments on five healthcare benchmarks demonstrate that FedAP achieves better accuracy compared to state-of-the-art methods (e.g., 10%+ accuracy improvement for PAMAP2) with faster convergence speed.

Vision Transformer (ViT) has achieved remarkable performance in computer vision. However, positional encoding in ViT makes it substantially difficult to learn the intrinsic equivariance in data. Ini- tial attempts have been made on designing equiv- ariant ViT but are proved defective in some cases in this paper. To address this issue, we design a Group Equivariant Vision Transformer (GE-ViT) via a novel, effective positional encoding opera- tor. We prove that GE-ViT meets all the theoreti- cal requirements of an equivariant neural network. Comprehensive experiments are conducted on standard benchmark datasets, demonstrating that GE-ViT significantly outperforms non-equivariant self-attention networks. The code is available at https://github.com/ZJUCDSYangKaifan/GEVit.

Superconductivity allows electrical current to flow without any energy loss, and thus making solids superconducting is a grand goal of physics, material science, and electrical engineering. More than 16 Nobel Laureates have been awarded for their contribution in superconductivity research. Superconductors are valuable for sustainable development goals (SDGs), such as climate change mitigation, affordable and clean energy, industry, innovation and infrastructure, and so on. However, a unified physics theory explaining all superconductivity mechanism is still unknown. It is believed that superconductivity is microscopically due to not only molecular compositions but also the geometric crystal structure. Hence a new dataset, S2S, containing both crystal structures and superconducting critical temperature, is built upon SuperCon and Material Project. Based on this new dataset, we propose a novel model, S2SNet, which utilizes the attention mechanism for superconductivity prediction. To overcome the shortage of data, S2SNet is pre-trained on the whole Material Project dataset with Masked-Language Modeling (MLM). S2SNet makes a new state-of-the-art, with out-of-sample accuracy of 92% and Area Under Curve (AUC) of 0.92. To the best of our knowledge, S2SNet is the first work to predict superconductivity with only information of crystal structures. This work is beneficial to superconductivity discovery and further SDGs. The code and datasets are available at https://github.com/supercond/S2SNet

User representation is essential for providing high-quality commercial services in industry. In our business scenarios, we face the challenge of learning universal (general-purpose) user representation. The universal representation is expected to be informative, and can handle various types of real-world applications without fine-tuning (e.g., applicable for both user profiling and the recall process in advertising). It shows great advantages compared to the solution of training a specific model for each downstream application. Specifically, we attempt to improve universal user representation from two points of views. First, a contrastive self-supervised learning paradigm is presented to guide the representation model training. It provides a unified framework that allows for long-term or short-term interest representation learning in a data-driven manner. Moreover, a novel multi-interest extraction module is presented. The module introduces an interest dictionary to capture principal interests of the given user, and then generate his/her interest-oriented representations via behavior aggregation. Experimental results demonstrate the effectiveness and applicability of the learned user representations. Such an industrial solution has now been deployed in various real-world tasks.

Cross-lingual speech adaptation aims to solve the problem of leveraging multiple rich-resource languages to build models for a low-resource target language. Since the low-resource language has limited training data, speech recognition models can easily overfit. Adapter is a versatile module that can be plugged into Transformer for parameter-efficient learning. In this paper, we propose to use adapters for parameter-efficient cross-lingual speech adaptation. Based on our previous MetaAdapter that implicitly leverages adapters, we propose a novel algorithm called SimAdapter for explicitly learning knowledge from adapters. Our algorithms can be easily integrated into the Transformer structure. MetaAdapter leverages meta-learning to transfer the general knowledge from training data to the test language. SimAdapter aims to learn the similarities between the source and target languages during fine-tuning using the adapters. We conduct extensive experiments on five-low-resource languages in the Common Voice dataset. Results demonstrate that MetaAdapter and SimAdapter can reduce WER by 2.98% and 2.55% with only 2.5% and 15.5% of trainable parameters compared to the strong full-model fine-tuning baseline. Moreover, we show that these two novel algorithms can be integrated for better performance with up to 3.55% relative WER reduction.

Implicit feedback is widely used in collaborative filtering methods for sequential recommendation. It is well known that implicit feedback contains a large number of values that aremissing not at random(MNAR); and the missing data is a mixture of negative and unknown feedback, making it difficult to learn users’ negative preferences. Recent studies modeledexposure, a latent missingness variable which indicates whether an item is exposed to a user, to give each missing entry a confidence of being negative feedback. However, these studies use static models and ignore the information in temporal dependencies among items, which seems to be an essential underlying factor to subsequent missingness. To model and exploit the dynamics of missingness, we propose a latent variable named “user intent” to govern the temporal changes of item missingness, and a hidden Markov model to represent such a process. The resulting framework captures the dynamic item missingness and incorporates it into matrix factorization (MF) for recommendation. We further extend the proposed framework to capture the dynamic preference of users, which results in a unified framework that is able to model different evolution patterns of user intent and user preference. We also explore two types of constraints to achieve a more compact and interpretable representation ofuser intents. Experiments on real-world datasets demonstrate the superiority of our method against state-of-the-art recommender systems.

Time series has wide applications in the real world and is known to be difficult to forecast. Since its statistical properties change over time, its distribution also changes temporally, which will cause severe distribution shift problem to existing methods. However, it remains unexplored to model the time series in the distribution perspective. In this paper, we term this as Temporal Covariate Shift (TCS). This paper proposes Adaptive RNNs (AdaRNN) to tackle the TCS problem by building an adaptive model that generalizes well on the unseen test data. AdaRNN is sequentially composed of two novel algorithms. First, we propose Temporal Distribution Characterization to better characterize the distribution information in the TS. Second, we propose Temporal Distribution Matching to reduce the distribution mismatch in TS to learn the adaptive TS model. AdaRNN is a general framework with flexible distribution distances integrated. Experiments on human activity recognition, air quality prediction, and financial analysis show that AdaRNN outperforms the latest methods by a classification accuracy of 2.6% and significantly reduces the RMSE by 9.0%. We also show that the temporal distribution matching algorithm can be extended in Transformer structure to boost its performance.

Parkinson’s disease is the second most common neurodegenerative disorder, commonly affecting elderly people over the age of 65. As the cardinal manifestation, hypomimia, referred to as impairments in normal facial expressions, stays covert. Even some experienced doctors may miss these subtle changes, especially in a mild stage of this disease. The existing methods for hypomimia recognition are mainly dominated by statistical variable-based methods with the help of traditional machine learning algorithms. Despite the success of recognizing hypomimia, they show a limited accuracy and lack the capability of performing semantic analysis. Therefore, developing a computer-aided diagnostic method for semantically recognizing hypomimia is appealing. In this article, we propose a Semantic Feature based Hypomimia Recognition network , named SFHR-NET , to recognize hypomimia based on facial videos. First, a Semantic Feature Classifier (SF-C) is proposed to adaptively adjust feature maps salient to hypomimia, which leads the encoder and classifier to focus more on areas of hypomimia-interest. In SF-C, the progressive confidence strategy (PCS) ensures more reliable semantic features. Then, a two-stream framework is introduced to fuse the spatial data stream and temporal optical stream, which allows the encoder to semantically and progressively characterize the rigid process of hypomimia. Finally, to improve the interpretability of the model, Gradient-weighted Class Activation Mapping (Grad-CAM) is integrated to generate attention maps that cast our engineered features into hypomimia-interest regions. These highlighted regions provide visual explanations for decisions of our network. Experimental results based on real-world data demonstrate the effectiveness of our method in detecting hypomimia.

Recently, extensive researches have been proposed to address the UDA problem, which aims to learn transferrable models for the unlabeled target domain. Among them, the optimal transport is a promising metric to align the representations of the source and target domains. However, most existing works based on optimal transport ignore the intra-domain structure, only achieving coarse pair-wise matching. The target samples distributed near the edge of the clusters, or far from their corresponding class centers are easily to be misclassified by the decision boundary learned from the source domain. In this paper, we present Reliable Weighted Optimal Transport (RWOT) for unsupervised domain adaptation, including novel Shrinking Subspace Reliability (SSR) and weighted optimal transport strategy. Specifically, SSR exploits spatial prototypical information and intra-domain structure to dynamically measure the sample-level domain discrepancy across domains. Besides, the weighted optimal transport strategy based on SSR is exploited to achieve the precise-pair-wise optimal transport procedure, which reduces negative transfer brought by the samples near decision boundaries in the target domain. RWOT also equips with the discriminative centroid clustering exploitation strategy to learn transfer features. A thorough evaluation shows that RWOT outperforms existing state-of-the-art method on standard domain adaptation benchmarks.

Domain adaptation (DA) has achieved a resounding success to learn a good classifier by leveraging labeled data from a source domain to adapt to an unlabeled target domain. However, in a general setting when the target domain contains classes that are never observed in the source domain, namely in Open Set Domain Adaptation (OSDA), existing DA methods failed to work because of the interference of the extra unknown classes. This is a much more challenging problem, since it can easily result in negative transfer due to the mismatch between the unknown and known classes. Existing researches are susceptible to misclassification when target domain unknown samples in the feature space distributed near the decision boundary learned from the labeled source domain. To overcome this, we propose Joint Partial Optimal Transport (JPOT), fully utilizing information of not only the labeled source domain but also the discriminative representation of unknown class in the target domain. The proposed joint discriminative prototypical compactness loss can not only achieve intra-class compactness and inter-class separability, but also estimate the mean and variance of the unknown class through backpropagation, which remains intractable for previous methods due to the blindness about the structure of the unknown classes. To our best knowledge, this is the first optimal transport model for OSDA. Extensive experiments demonstrate that our proposed model can significantly boost the performance of open set domain adaptation on standard DA datasets.