ICML 2025 Papers

Knowledge distillation typically minimizes the Kullback–Leibler (KL) divergence between teacher and student logits. However, optimizing the KL divergence can be challenging for the student and often leads to sub-optimal solutions. We further show that gradients induced by KL divergence scale with the magnitude of the teacher logits, thereby diminishing updates on low-probability channels. This imbalance weakens the transfer of inter-class information and in turn limits the performance improvements achievable by the student. To mitigate this issue, we propose a plug-and-play auxiliary ranking loss based on Kendall’s $\tau$ coefficient that can be seamlessly integrated into any logit-based distillation framework. It supplies inter-class relational information while rebalancing gradients toward low-probability channels. We demonstrate that the proposed ranking loss is largely invariant to channel scaling and optimizes an objective aligned with that of KL divergence, making it a natural complement rather than a replacement. Extensive experiments on CIFAR-100, ImageNet, and COCO datasets, as well as various CNN and ViT teacher-student architecture combinations, demonstrate that our plug-and-play ranking loss consistently boosts the performance of multiple distillation baselines.

Neural Operators that directly learn mappings between function spaces, such as Deep Operator Networks (DONs) and Fourier Neural Operators (FNOs), have received considerable attention. Despite the universal approximation guarantees for DONs and FNOs, there is currently no optimization convergence guarantee for learning such networks using gradient descent (GD). In this paper, we address this open problem by presenting a unified framework for optimization based on GD and applying it to establish convergence guarantees for both DONs and FNOs. In particular, we show that the losses associated with both of these neural operators satisfy two conditions—restricted strong convexity (RSC) and smoothness—that guarantee a decrease on their loss values due to GD. Remarkably, these two conditions are satisfied for each neural operator due to different reasons associated with the architectural differences of the respective models. One takeaway that emerges from the theory is that wider networks benefit optimization convergence guarantees for both DONs and FNOs. We present empirical results on canonical operator learning problems to support our theoretical results and find that larger widths benefit training.

Diffusion probabilistic models (DPMs) have recently demonstrated impressive generative capabilities. There is emerging evidence that their sample reconstruction ability can yield meaningful representations for recognition tasks. In this paper, we demonstrate that the objectives underlying generation and representation learning are not perfectly aligned. Through a spectral analysis, we find that minimizing the mean squared error (MSE) between the original graph and its reconstructed counterpart does not necessarily optimize representations for downstream tasks. Instead, focusing on reconstructing a small subset of features, specifically those capturing global information, proves to be more effective for learning powerful representations. Motivated by these insights, we propose a novel framework, the Smooth Diffusion Model for Graphs (SDMG), which introduces a multi-scale smoothing loss and low-frequency information encoders to promote the recovery of global, low-frequency details, while suppressing irrelevant high-frequency noise. Extensive experiments validate the effectiveness of our method, suggesting a promising direction for advancing diffusion models in graph representation learning.

Real-world datasets often follow a long-tailed distribution, making generalization to tail classes difficult. Recent methods resorted to long-tail variants of Sharpness-Aware Minimization (SAM), such as ImbSAM and CC-SAM, to improve generalization by flattening the loss landscape. However, these attempts face a trade-off between computational efficiency and control over the loss landscape. On the one hand, ImbSAM is efficient but offers only coarse control as it excludes head classes from the SAM process. On the other hand, CC-SAM provides fine-grained control through class-dependent perturbations but at the cost of efficiency due to multiple backpropagations. Seeing this dilemma, we introduce Focal-SAM, which assigns different penalties to class-wise sharpness, achieving fine-grained control without extra backpropagations, thus maintaining efficiency. Furthermore, we theoretically analyze Focal-SAM's generalization ability and derive a sharper generalization bound. Extensive experiments on both traditional and foundation models validate the effectiveness of Focal-SAM.

The peer review process in major artificial intelligence (AI) conferences faces unprecedented challenges with the surge of paper submissions (exceeding 10,000 submissions per venue), accompanied by growing concerns over review quality and reviewer responsibility. This position paper argues for **the need to transform the traditional one-way review system into a bi-directional feedback loop where authors evaluate review quality and reviewers earn formal accreditation, creating an accountability framework that promotes a sustainable, high-quality peer review system.** The current review system can be viewed as an interaction between three parties: the authors, reviewers, and system (i.e., conference), where we posit that all three parties share responsibility for the current problems. However, issues with authors can only be addressed through policy enforcement and detection tools, and ethical concerns can only be corrected through self-reflection. As such, this paper focuses on reforming reviewer accountability with systematic rewards through two key mechanisms: (1) a two-stage bi-directional review system that allows authors to evaluate reviews while minimizing retaliatory behavior, (2) a systematic reviewer reward system that incentivizes quality reviewing. We ask for the community's strong interest in these problems and the reforms that are needed to enhance the peer review process.

For data with intra-class Asymmetric instance Distribution or Multiple High-density Clusters (ADMHC), outliers are real and have specific patterns for data classification, where the class body is necessary and difficult to identify. Previous Feature Selection (FS) methods score features based on all training instances or rarely target intra-class ADMHC. In this paper, we propose a supervised FS method, Stray Intrusive Outliers-based FS (SIOFS), for data classification with intra-class ADMHC. By focusing on Stray Intrusive Outliers (SIOs), SIOFS modifies the skewness coefficient and fuses the threshold in the 3$\sigma$ principle to identify the class body, scoring features based on the intrusion degree of SIOs. In addition, the refined density-mean center is proposed to represent the general characteristics of the class body reasonably. Mathematical formulations, proofs, and logical exposition ensure the rationality and universality of the settings in the proposed SIOFS method. Extensive experiments on 16 diverse benchmark datasets demonstrate the superiority of SIOFS over 12 state-of-the-art FS methods in terms of classification accuracy, normalized mutual information, and confusion matrix. SIOFS source codes is available at https://github.com/XXXly/2025-ICML-SIOFS

Self-supervised learning (SSL) has proven to be a powerful approach for extracting biologically meaningful representations from single-cell data. To advance our understanding of SSL methods applied to single-cell data, we present scSSL-Bench, a comprehensive benchmark that evaluates nineteen SSL methods. Our evaluation spans nine datasets and focuses on three common downstream tasks: batch correction, cell type annotation, and missing modality prediction. Furthermore, we systematically assess various data augmentation strategies. Our analysis reveals task-specific trade-offs: the specialized single-cell frameworks, scVI, CLAIRE, and the finetuned scGPT excel at uni-modal batch correction, while generic SSL methods, such as VICReg and SimCLR, demonstrate superior performance in cell typing and multi-modal data integration. Random masking emerges as the most effective augmentation technique across all tasks, surpassing domain-specific augmentations. Notably, our results indicate the need for a specialized single-cell multi-modal data integration framework. scSSL-Bench provides a standardized evaluation platform and concrete recommendations for applying SSL to single-cell analysis, advancing the convergence of deep learning and single-cell genomics.

Heterogeneous Graph Neural Networks (HGNNs), have demonstrated excellent capabilities in processing heterogeneous information networks. Self-supervised learning on heterogeneous graphs, especially contrastive self-supervised strategy, shows great potential when there are no labels. However, this approach requires the use of carefully designed graph augmentation strategies and the selection of positive and negative samples. Determining the exact level of similarity between sample pairs is non-trivial.To solve this problem, we propose a novel self-supervised Heterogeneous graph neural network with Optimal Transport (HGOT) method which is designed to facilitate self-supervised learning for heterogeneous graphs without graph augmentation strategies. Different from traditional contrastive self-supervised learning, HGOT employs the optimal transport mechanism to relieve the laborious sampling process of positive and negative samples. Specifically, we design an aggregating view (central view) to integrate the semantic information contained in the views represented by different meta-paths (branch views). Then, we introduce an optimal transport plan to identify the transport relationship between the semantics contained in the branch view and the central view. This allows the optimal transport plan between graphs to align with the representations, forcing the encoder to learn node representations that are more similar to the graph space and of higher quality. Extensive experiments on four real-world datasets demonstrate that our proposed HGOT model can achieve state-of-the-art performance on various downstream tasks. In particular, in the node classification task, HGOT achieves an average of more than 6\% improvement in accuracy compared with state-of-the-art methods.

Self-supervised learning (SSL) methods via joint embedding architectures have proven remarkably effective at capturing semantically rich representations with strong clustering properties, magically in the absence of label supervision. Despite this, few of them have explored leveraging these untapped properties to improve themselves. In this paper, we provide an evidence through various metrics that the encoder's output *encoding* exhibits superior and more stable clustering properties compared to other components. Building on this insight, we propose a novel positive-feedback SSL method, termed **Re**presentation **S**elf-**A**ssignment (ReSA), which leverages the model's clustering properties to promote learning in a self-guided manner. Extensive experiments on standard SSL benchmarks reveal that models pretrained with ReSA outperform other state-of-the-art SSL methods by a significant margin. Finally, we analyze how ReSA facilitates better clustering properties, demonstrating that it effectively enhances clustering performance at both fine-grained and coarse-grained levels, shaping representations that are inherently more structured and semantically meaningful.

We introduce Random Feature Representation Boosting (RFRBoost), a novel method for constructing deep residual random feature neural networks (RFNNs) using boosting theory. RFRBoost uses random features at each layer to learn the functional gradient of the network representation, enhancing performance while preserving the convex optimization benefits of RFNNs. In the case of MSE loss, we obtain closed-form solutions to greedy layer-wise boosting with random features. For general loss functions, we show that fitting random feature residual blocks reduces to solving a quadratically constrained least squares problem. Through extensive numerical experiments on tabular datasets for both regression and classification, we show that RFRBoost significantly outperforms RFNNs and end-to-end trained MLP ResNets in the small- to medium-scale regime where RFNNs are typically applied. Moreover, RFRBoost offers substantial computational benefits, and theoretical guarantees stemming from boosting theory.

Although Recommender Systems (RS) have been well-developed for various fields of applications, they often suffer from a crisis of platform credibility with respect to RS confidence and fairness, which may drive users away, threatening the platform's long-term success. In recent years, some works have tried to solve these issues; however, they lack strong statistical guarantees. Therefore, there is an urgent need to solve both issues with a unifying framework with robust statistical guarantees. In this paper, we propose a novel and reliable framework called Equitable and Statistically Unbiased Recommendation (ENSUR)) to dynamically generate prediction sets for users across various groups, which are guaranteed 1) to include ground-truth items with user-predefined high confidence/probability (e.g., 90\%); 2) to ensure user fairness across different groups; 3) to have minimum efficient average prediction set sizes. We further design an efficient algorithm named Guaranteed User Fairness Algorithm (GUFA) to optimize the proposed method and derive upper bounds of risk and fairness metrics to speed up the optimization process. Moreover, we provide rigorous theoretical analysis concerning risk and fairness control and minimum set size. Extensive experiments validate the effectiveness of the proposed framework, which aligns with our theoretical analysis.

Electromagnetic structure (EMS) design plays a critical role in developing advanced antennas and materials, but remains challenging due to high-dimensional design spaces and expensive evaluations. While existing methods commonly employ high-quality predictors or generators to alleviate evaluations, they are often data-intensive and struggle with real-world scale and budget constraints. To address this, we propose a novel method called Progressive Quadtree-based Search (PQS). Rather than exhaustively exploring the high-dimensional space, PQS converts the conventional image-like layout into a quadtree-based hierarchical representation, enabling a progressive search from global patterns to local details. Furthermore, to lessen reliance on highly accurate predictors, we introduce a consistency-driven sample selection mechanism. This mechanism quantifies the reliability of predictions, balancing exploitation and exploration when selecting candidate designs. We evaluate PQS on two real-world engineering tasks, i.e., Dual-layer Frequency Selective Surface and High-gain Antenna. Experimental results show that our method can achieve satisfactory designs under limited computational budgets, outperforming baseline methods. In particular, compared to generative approaches, it cuts evaluation costs by 75∼85%, effectively saving 20.27∼38.80 days of product designing cycle.

Label distribution learning (LDL) is a novel learning paradigm that emulates label polysemy by assigning label distributions over the label space. However, recent LDL work seems to exhibit a notable contradiction: 1) existing LDL methods employ auxiliary tasks to enhance performance, which narrows their focus to specific applications, thereby lacking generalizability; 2) conversely, LDL methods without auxiliary tasks rely on losses tailored solely to the primary task, lacking beneficial data to guide the learning process. In this paper, we propose S-LDL, a novel and minimalist solution that generates subtask label distributions, i.e., a form of extra supervised information, to reconcile the above contradiction. S-LDL encompasses two key aspects: 1) an algorithm capable of generating subtasks without any prior/expert knowledge; and 2) a plug-andplay framework seamlessly compatible with existing LDL methods, and even adaptable to derivative tasks of LDL. Our analysis and experiments demonstrate that S-LDL is effective and efficient. To the best of our knowledge, this paper represents the first endeavor to address LDL via subtasks.

Detecting anomalies with limited supervision is challenging due to the scarcity of labeled anomalies, which often fail to capture the diversity of abnormal behaviors. We propose Weakly Supervised Anomaly Detection via Dual-Tailed Kernel (WSAD-DT), a novel framework that learns robust latent representations to distinctly separate anomalies from normal samples under weak supervision. WSAD-DT introduces two centroids—one for normal samples and one for anomalies—and leverages a dual-tailed kernel scheme: a light-tailed kernel to compactly model in-class points and a heavy-tailed kernel to main- tain a wider margin against out-of-class instances. To preserve intra-class diversity, WSAD-DT in- corporates kernel-based regularization, encouraging richer representations within each class. Furthermore, we devise an ensemble strategy that partition unlabeled data into diverse subsets, while sharing the limited labeled anomalies among these partitions to maximize their impact. Empirically, WSAD-DT achieves state-of-the-art performance on several challenging anomaly detection benchmarks, outperforming leading ensemble-based methods such as XGBOD.

Due to its simplicity, effectiveness and robustness, naive Bayes (NB) has continued to be one of the top 10 data mining algorithms. To improve its performance, a large number of improved algorithms have been proposed in the last few decades. However, in addition to Gaussian naive Bayes (GNB), there is little work on numerical attributes. At the same time, none of them takes into account the correlations among instances. To fill this gap, we propose a novel algorithm called instance correlation graph-based naive Bayes (ICGNB). Specifically, it first uses original attributes to construct an instance correlation graph (ICG) to represent the correlations among instances. Then, it employs a variational graph auto-encoder (VGAE) to generate new attributes from the constructed ICG and uses them to augment original attributes. Finally, it weights each augmented attribute to alleviate the attribute redundancy and builds GNB on the weighted attributes. The experimental results on tens of datasets show that ICGNB significantly outperforms its deserved competitors.Our codes and datasets are available at https://github.com/jiangliangxiao/ICGNB.

Bipartite ranking is a fundamental supervised learning problem, with the goal of learning a ranking over instances with maximal area under the ROC curve (AUC) against a single binary target label. However, one may often observe multiple binary target labels, e.g., from distinct human annotators. How can one synthesize such labels into a single coherent ranking? In this work, we formally analyze two approaches to this problem—loss aggregation and label aggregation—by characterizing their Bayes-optimal solutions. We show that while both approaches can yield Pareto-optimal solutions, loss aggregation can exhibit label dictatorship: one can inadvertently (and undesirably) favor one label over others. This suggests that label aggregation can be preferable to loss aggregation, which we empirically verify.

We study active data collection strategies for operator learning when the target operator is linear and the input functions are drawn from a mean-zero stochastic process with continuous covariance kernels. With an active data collection strategy, we establish an error convergence rate in terms of the decay rate of the eigenvalues of the covariance kernel. We can achieve arbitrarily fast error convergence rates with sufficiently rapid eigenvalue decay of the covariance kernels. This contrasts with the passive (i.i.d.) data collection strategies, where the convergence rate is never faster than linear decay ($\sim n^{-1}$). In fact, for our setting, we show a \emph{non-vanishing} lower bound for any passive data collection strategy, regardless of the eigenvalues decay rate of the covariance kernel. Overall, our results show the benefit of active data collection strategies in operator learning over their passive counterparts.

Over 40\% of the global population lives within 100 kilometers of the coast, which contributes more than \$8 trillion annually to the global economy. Unfortunately, coastal ecosystems are increasingly vulnerable to more frequent and intense extreme weather events and rising sea levels. Coastal scientists use numerical models to simulate complex physical processes, but these models are often slow and expensive. In recent years, deep learning has become a promising alternative to reduce the cost of numerical models. However, progress has been hindered by the lack of a large-scale, high-resolution coastal simulation dataset to train and validate deep learning models. Existing studies often focus on relatively small datasets and simple processes. To fill this gap, we introduce a decade-long, high-resolution (<100m) coastal circulation modeling dataset on a real-world 3D mesh in southwest Florida with around 6 million cells. The dataset contains key oceanography variables (e.g., current velocities, free surface level, temperature, salinity) alongside external atmospheric and river forcings. We evaluated a customized Vision Transformer model that takes initial and boundary conditions and external forcings and predicts ocean variables at varying lead times. The dataset provides an opportunity to benchmark novel deep learning models for high-resolution coastal simulations (e.g., physics-informed machine learning, neural operator learning). The code and dataset can be accessed at https://github.com/spatialdatasciencegroup/CoastalBench.

This paper studies the long-tailed semi-supervised learning (LTSSL) with distribution mismatch, where the class distribution of the labeled training data follows a long-tailed distribution and mismatches with that of the unlabeled training data. Most existing methods introduce auxiliary classifiers (experts) to model various unlabeled data distributions and produce pseudo-labels, but the expertises of various experts are not fully utilized. We observe that different experts are good at predicting different intervals of samples, e.g., long-tailed expert is skilled in samples located in the head interval and uniform expert excels in samples located in the medium interval. Therefore, we propose a dynamic expert assignment module that can estimate the class membership (i.e., head, medium, or tail class) of samples, and dynamically assigns suitable expert to each sample based on the estimated membership to produce high-quality pseudo-label in the training phase and produce prediction in the testing phase. We also theoretically reveal that integrating different experts' strengths will lead to a smaller generalization error bound. Moreover, we find that the deeper features are more biased toward the head class but with more discriminative ability, while the shallower features are less biased but also with less discriminative ability. We, therefore, propose a multi-depth feature fusion module to utilize different depth features to mitigate the model bias. Our method demonstrates its effectiveness through comprehensive experiments on the CIFAR-10-LT, STL-10-LT, and SVHN-LT datasets across various settings. The code is available at https://github.com/yaxinhou/Meta-Expert.