论文检索

Social Aspects · Accountability, Transparency, and Interpretability

Gouki Minegishi, Jingyuan Feng, Hiroki Furuta, Takeshi Kojima, Yusuke Iwasawa, Yutaka Matsuo

Analogy is a central faculty of human intelligence, enabling abstract patterns discovered in one domain to be applied to another. However, the mechanisms underlying analogical reasoning in Transformers remain poorly understood. In this work, inspired by the notion of functors in category theory, we formalize analogical reasoning as the inference of correspondences between entities across categories. Based on this formulation, we introduce synthetic tasks that evaluate the emergence of analogical reasoning under controlled settings. We find that the emergence of analogical reasoning is highly sensitive to data characteristics, optimization choices, and model scale. Through mechanistic analysis, we show that analogical reasoning in Transformers decomposes into two key components: (1) geometric alignment of relational structure in the embedding space, and (2) the application of a functor within the Transformer. These mechanisms enable models to transfer relational structure from one category to another, realizing analogy. Finally, we quantify these effects and find that the same trends are observed in pretrained LLMs. In doing so, we move analogy from an abstract cognitive notion to a concrete, mechanistically grounded phenomenon in modern neural networks.

General Machine Learning · Representation Learning

Yuchen Guan, Xiao Li, Zongyu Guo, Xiaoyi Zhang, Xiulian Peng, Chun Yuan, Yan Lu

We propose a new paradigm for long video understanding by treating a long video as a Neural Knowledge Representation (NKR). NKR represent video contents neither as a stream of tokens or pre-organized databases, but as an individual small portion of network weights attached to the VLM backbone. The NKR weights is optimized to encapsulate the video's semantic content via a novel Agentic Knowledge Distillation (AKD) process, where an agent automatically synthesizes dense descriptions and question-answer pairs to distill the video's knowledge into the NKR. While AKD serves as a comprehensive, one-time encoding phase, the resulting NKR transforms the video into a portable, reusable asset. At inference, the lightweight NKR is mounted onto a frozen Vision-Language Model (VLM), enabling direct, query-based understanding without reloading or re-encoding the the original video. This approach decouples video length from inference cost, offering high amortized efficiency for multi-turn video understanding. Experiments on the LVBench benchmark show our method achieves performance comparable to state-of-the-art approaches while reducing end-to-end latency by over two orders of magnitude, opening new possibilities for interactive long-video understanding.

Deep Learning · Foundation Models

Yi Fang, Haoran Xu, Jiaxin Han, Sirui Ding, Yizhi Wang, Yue Wang, Xuan Wang

Foundation models have revolutionized AI, yet biological applications often repurpose general architectures without accounting for the intrinsic structural and functional properties of distinct modalities, such as genomic and proteomic sequences. Consequently, these architectures lack the inductive biases required to capture the complex ``grammars" inherent to biological data, resulting in suboptimal performance. To address this, we introduce BioArc, a framework utilizing Neural Architecture Search (NAS) to shift from intuition-driven design to automated data-driven discovery. Unlike standard NAS restricted to homogeneous spaces, BioArc navigates a heterogeneous space for open-ended composition of architectural blocks. By systematically analyzing the interplay between architecture, tokenization, and training across modalities, BioArc identifies novel hybrid architectures that surpass state-of-the-art models while being up to 25x smaller. We distill these findings into empirical design principles and validate their biological relevance, demonstrating how our designs hierarchically capture the underlying biological grammar. Additionally, we introduce an agentic framework to predict optimal architectures for new tasks. Overall, BioArc provides a data-driven methodology for developing the next generation of efficient biological foundation models and task-specific networks.

Deep Learning · Graph Neural Networks

Adrien Lagesse, Marc Lelarge

We propose a novel benchmarking methodology for graph neural networks (GNNs) based on the graph alignment problem, a combinatorial optimization task that generalizes graph isomorphism by aligning two unlabeled graphs to maximize overlapping edges. We frame this problem as a self-supervised learning task and present several methods to generate graph alignment datasets using synthetic random graphs and real-world graph datasets from multiple domains. For a given graph dataset, we generate a family of graph alignment datasets with increasing difficulty, allowing us to rank the performance of various architectures. Our experiments prove that there is an optimal task difficulty for having a statistically relevant ranking of different models and that, even on a structure-only task, anisotropic models perform better compared to isotropic ones. To further prove that our synthetic task capture meaningful information, we show its effectiveness for self-supervised GNN pre-training: the learned node embeddings can be leveraged as positional encodings by transformers for graph regression or can be used to reconstruct the full structure of the graph with 98% accuracy. To support reproducibility and further research, we provide an open-source Python package to generate graph alignment datasets and benchmark new GNN architectures.

Deep Learning · Graph Neural Networks

Chendi Qian, Christopher Morris

Semidefinite programs (SDPs) are a powerful framework for convex optimization and for constructing strong relaxations of hard combinatorial problems. However, solving large SDPs can be computationally expensive, motivating the use of machine learning models as fast computational surrogates. Graph neural networks (GNNs) are a natural candidate in this setting due to their sparsity-awareness and ability to model variable-constraint interactions. In this work, we study what expressive power is sufficient to recover optimal SDP solutions. We first prove negative results showing that standard GNN architectures fail on recovering linear SDP solutions. We then identify a more expressive architecture that captures the key structure of SDPs and can, in particular, emulate the updates of a standard first-order solver. Empirically, on both synthetic and SDPLIB benchmarks of various classes of SDPs, this more expressive architecture achieves consistently lower prediction error and objective gap than theoretically weaker baselines. Finally, using the learned high-quality predictions to warm-start the first-order solver yields practical speedups of up to 80%.

Applications · Computer Vision

Luke Miller, Yugyung Lee

Segmenting small and sparse structures in large-scale images is fundamentally constrained by pixel-level, lattice-bound computation and extreme class imbalance--dense, full-resolution inference scales poorly and forces most pipelines to rely on fixed regionization or downsampling, coupling computational cost to image resolution and attenuating boundary evidence precisely where minority structures are most informative. We introduce **SEMIR** (*Semantic Minor-Induced Representation Learning*), a representation framework that decouples inference from the native grid by learning a task-adapted inference space. **SEMIR** transforms the underlying grid graph into a compact, boundary-aligned graph minor through parameterized edge contraction, node deletion, and edge deletion, while preserving an exact lifting map from minor predictions to lattice labels. Minor construction is formalized as a few-shot optimization problem that replaces hand-tuned preprocessing with a *boundary-alignment objective*: minor parameters are learned by maximizing agreement between predicted boundary elements and class-agnostic semantic edges under a *boundary Dice criterion*, and the induced minor is annotated with scale- and rotation-robust geometric and intensity descriptors and supports efficient region-level inference via message passing on a graph neural network (GNN) with relational edge features. We benchmark **SEMIR** on three tumor segmentation datasets—**BraTS2021**, **KiTS2023**, and **LiTS2017**—where targets exhibit high structural variability and distributional uncertainty, providing a stringent testbed for *structure-adaptive inference*. **SEMIR** yields consistent improvements in *minority-structure Dice* at practical runtime, positioning *minor-induced representations* as a principled alternative to pixel-centric segmentation in challenging, high-variability visual domains.

Deep Learning · Graph Neural Networks

Yao Cheng, Siqiang Luo

Relational deep learning (RDL) converts relational databases (RDBs) into heterogeneous graphs, but graphs derived directly from database schemas are often not well suited for how graph neural networks (GNNs) perform relational reasoning. We study what makes a relational graph suitable for deep learning and show that schema-derived graphs suffer from two systematic failures: information overload and semantic fragmentation. Through an empirical analysis on real-world databases, we find that effective graphs arise from a task-dependent balance between removing task-irrelevant structure and injecting task-aligned relational connectivity. Filtering exhibits a non-monotonic effect on performance, while structural injection is beneficial only when it reflects the logic of the downstream task. Based on these findings, we develop an end-to-end structural optimizer that applies both operations to adapt relational graphs automatically. Across 23 tasks spanning classification, regression, and recommendation, the optimized graphs consistently improve accuracy while often reducing inference cost. Code and data are available at https://anonymous.4open.science/r/Structural_Optimizer_RDL-0F74/.

Deep Learning · Graph Neural Networks

Jialiang Wang, Hanmo Liu, Shimin Di, Zhili Wang, Jiachuan Wang, Lei Chen, Xiaofang Zhou

Designing high-performance neural networks for new tasks requires balancing optimization quality with search efficiency. Current methods fail to achieve this balance: neural architectural search is computationally expensive, while model retrieval often yields suboptimal static checkpoints. To resolve this dilemma, we model the performance gains induced by fine-grained architectural modifications as edit-effect evidence and build evidence graphs from prior tasks. By constructing a retrieval-augmented model refinement framework, our proposed M-DESIGN dynamically weaves historical evidence to discover near-optimal modification paths. M-DESIGN features an adaptive retrieval mechanism that quickly calibrates the evolving transferability of edit-effect evidence from different sources. To handle out-of-distribution shifts, we introduce predictive task planners that extrapolate gains from multi-hop evidence, thereby reducing reliance on an exhaustive repository. Based on our model knowledge base of 67,760 graph neural networks across 22 datasets, extensive experiments demonstrate that M-DESIGN consistently outperforms baselines, achieving the search-space best performance in 26 out of 33 cases under a strict budget. Code and data are available at: https://anonymous.4open.science/r/M-DESIGN-245/.

Social Aspects · Privacy

Sujia Huang, Lele Fu, Shunxin Xiao, Xiaoya Zhang, Chunyan Xu, Tong Zhang, Bo Huang, Zhen Cui

Federated Graph Learning (FGL) facilitates privacy-preserving collaborative training of graph neural networks, yet homophily heterogeneity across subgraphs triggers optimization conflicts that degrade model generalization. Most existing solutions rely on multi-channel architectures to mitigate such conflict, which increase the burden on edge devices and lack theoretical convergence guarantees. To overcome these limitations, we propose FedGCM, a novel FGL framework with Group-oriented Conflict Mitigation, which aligns inconsistent optimization objectives via a tailored gradient surgery scheme. Specifically, FedGCM first divides clients into distinct groups based on their homophily levels, a strategy that precludes exhaustive client-to-client conflict assessments. To resolve inter-group interference, we develop RPGrad, a gradient surgery mechanism based on residual projection, which integrates synergistic knowledge while filtering inter-group conflicts. The refined updates are then transmitted in a group-wise fashion, effectively alleviating optimization conflicts induced by homophily heterogeneity without augmenting the client-side burden. Furthermore, we provide a formal theoretical analysis establishing the convergence of FedGCM. Extensive experiments on both homophilous and heterophilous graphs demonstrate that FedGCM consistently achieves advanced performance.

Optimization · Everything Else

Minhao Zou, Tao Ren, Jinyang Jiang, Rui Tao, Zehao Li, Jiale Fu, Hui Shao, Xianhua Liu, Yijie Peng

Gradient-based optimization is fundamental to deep learning, yet standard backpropagation (BP) is inherently limited by the requirement of differentiability, rendering it brittle when encountering piecewise-constant objectives with vanishing gradients (e.g., hard 0-1 loss) or black-box feedback. While likelihood ratio (LR) methods offer a theoretical alternative, their high variance in high-dimensional spaces often undermines training stability and scalability. We propose OVLR (Output-Level Variance-Reduced Likelihood Ratio), a simple yet powerful framework that circumvents this fundamental trade-off by providing a unified solution for efficient, scalable, and robust gradient estimation. OVLR achieves dramatic variance reduction by performing perturbations and antithetic sampling in the low-dimensional output space. Crucially, the method maintains high computational efficiency: it requires only a single deterministic forward pass through the neural network, with additional costs restricted to evaluating the loss function across multiple samples. Designed as a drop-in replacement, OVLR integrates seamlessly into automatic differentiation frameworks via vector-Jacobian products, enabling the direct optimization of objectives with vanishing or pathological gradients, such as the 0-1 loss for noise-tolerant classification and truncated losses for outlier-resistant regression, where BP fails to provide reliable learning signals. Extensive empirical results across image classification, generative modeling, language modeling, and robot imitation learning demonstrate that OVLR not only matches BP performance on problems with informative gradients, but also provides a decisive advantage on problems with vanishing or inaccessible gradients.

General Machine Learning · Transfer, Multitask and Meta-learning

Zhiyuan Yu, Enneng Yang, Hao Jiang, Guojie Zhu, Feihong He, Peng Wang, Li Shen

Upcycling, a strategy that initializes Mixture-of-Experts (MoE) by replicating pre-trained feed-forward or MoE networks to expand model capacity, has become a popular method in continual learning due to its effectiveness in mitigating catastrophic forgetting. However, existing paradigms rely on indiscriminate expansion prioritize performance at the cost of extreme inefficiency, introducing parameter redundancy without exploiting the structural heterogeneity essential for counteracting forgetting with architectural economy. To address this, we investigate the determinants of forgetting in training dynamics using Centered Kernel Alignment (CKA) and loss landscape flatness to analyze the behavior of pre- and post-expansion MoE layers, uncovering instability in deep-layer representations and heterogeneous expert sensitivity to new tasks, thereby demonstrating the potential of selective upcycling to eliminate redundancy. Consequently, we propose a dynamic bilevel optimization framework to guide adaptive upcycling, featuring an outer loop employing a Gumbel-Softmax differentiable mask to perform Neural Architecture Search (NAS) for adaptive growth, while an inner loop optimizes weight updates via task objectives and CKA-regularized replay. Experiments on TRACE benchmark demonstrate that our proposed method achieves better average accuracy with 80\% forgetting reduction, while effectively eliminating 60\% of redundant parameter expansion that standard upcycling would introduce.

Theory · Optimization

Anming Gu, Juno Kim

The mean-field Langevin dynamics (MFLD) minimizes an entropy-regularized nonlinear convex functional on the Wasserstein space over $\mathbb{R}^d$, and has gained attention recently as a model for the gradient descent dynamics of interacting particle systems such as infinite-width two-layer neural networks. However, many problems of interest have constrained domains, which are not solved by existing mean-field algorithms due to the global diffusion term. We study the optimization of probability measures constrained to a convex subset of $\mathbb{R}^d$ by proposing the \emph{mirror mean-field Langevin dynamics} (MMFLD), an extension of MFLD to the mirror Langevin framework. We obtain linear convergence guarantees for the continuous MMFLD via a uniform log-Sobolev inequality, and uniform-in-time propagation of chaos results for its time- and particle-discretized counterpart.

Deep Learning · Graph Neural Networks

Soyoung Park, Hwanjun Song, Sungsu Lim

Uncertainty quantification (UQ) in graph neural networks (GNNs) is crucial in high-stakes domains but remains a significant challenge. In graph settings, message passing often relies on strong assumptions such as exchangeability, which are rarely satisfied in practice. Moreover, achieving reliable UQ typically requires costly resampling or post-hoc calibration. To address these issues, we introduce Quantile-free Prediction Interval GNN (QpiGNN), a framework that builds on quantile regression (QR) to enable GNN-based UQ by directly optimizing coverage and interval width without requiring quantile inputs or post-processing. QpiGNN employs a dual-head architecture that decouples prediction and uncertainty, and is trained with label-only supervision through a quantile-free joint loss. This design allows efficient training and yields robust prediction intervals, with theoretical guarantees of asymptotic coverage and near-optimal width under mild assumptions. Experiments on 19 synthetic and real-world benchmarks show QpiGNN achieves average 22% higher coverage and 50% narrower intervals than baselines, while ensuring efficiency and robustness to noise and structural shifts.

Applications · Neuroscience, Cognitive Science

Junfeng Lu, Zijie Yu, SHAOYANG CUI, Gan He, Ruiqin Xiong, Kai Du, Tiejun Huang

Biophysically detailed neural networks represent a promising frontier for brain-inspired AI, offering intrinsic spatio-temporal dynamics to enhance the expressivity and computational density of deep learning systems. However, general-purpose deep learning frameworks suffer from a fundamental mismatch between their dense parallel optimizations and the irregular, tree-structured complexity of biological mechanisms. In this work, we propose **HelioX**, a **GPU-native** framework designed to unify high-performance simulation with scalable training. Unlike approaches that adapt biology to existing deep learning tools, HelioX adopts a "GPU-to-Biophysics" paradigm. We tailor the underlying GPU parallelism to biological structures by implementing custom-fused CUDA kernels for both the Dendritic Hierarchical Scheduling (DHS) algorithm and its gradient propagation. This design eliminates the runtime overhead of generic automatic differentiation and enables multi-stream concurrency for spike generation and equation assembly. Experimental results demonstrate that HelioX outperforms standard simulators (NEURON) by orders of magnitude and surpasses prior GPU-based solvers in both speed and scalability. We successfully train deep biophysical MLPs and organism-scale biophysical neural networks (e.g., the BAAIWorm C. elegans model) on a single consumer-grade GPU. HelioX establishes a new standard for computational efficiency, enabling the training of biophysically detailed models at scales previously unattainable.

Deep Learning · Other Representation Learning

Kry Yik Chau Lui, Cheng Chi, Kishore Basu, Yanshuai Cao

Deep neural networks excel in vision, language, and audio, yet continue to underperform relative to tree-based models on tabular data. We identify and extend inductive biases crucial for tabular learning—robustness to irrelevant features, axis alignment, localized irregularities, feature heterogeneity, and training stability—and propose *LassoFlexNet*, a novel architecture coupled with a new training algorithm. LassoFlexNet employs a Tied Group Lasso mechanism that sparsely selects raw inputs based on nonlinear per-feature embeddings. This design encourages a raw input variable to contribute jointly with others only if it provides marginal predictive value, linearly or nonlinearly. The resulting non-homogeneity and localized irregularities introduce optimization challenges that defeat standard stochastic and proximal-gradient methods. To address this, we develop a *Sequential Hierarchical Proximal Gradient optimizer with exponential moving averages (EMA)*, enabling stable training. Across 52 datasets from three recent benchmarks, LassoFlexNet matches or surpasses state-of-the-art tree-based models, achieving up to 10% relative gains while improving interpretability. We further validate our design through ablation studies and prove enhanced expressivity for a key architectural component.

Deep Learning · Graph Neural Networks

Xianlin Zeng, Fan Xia, Xiangyu Chen

Text-attributed Graphs (TAGs) incorporate textual node attributes with graph structures to describe rich relational semantics. Recent efforts to integrate Graph Neural Networks (GNNs) and Large Language Models (LLMs) have shown promise for learning on TAGs, yet achieving well-aligned representations remains challenging. Prior studies largely rely on heuristics that perform coarse-grained matching. They lack sufficient constraints and ignore distributional alignment, leading to representation drift and limited generalization. Building on Energy-based Models (EBMs), we propose an **E**nergy-based **R**epresentation **Align**ment (ERAlign) framework that projects GNN-encoded graph structure and LLM-derived text embeddings in a shared latent space to achieve distribution consistency. Concretely, layer-wise alignment is quantified by a distance metric and optimized via an EBM objective. By decreasing energy values, our framework yields well-aligned representations for downstream tasks. During training, we introduce Energy Discrepancy (ED) to avoid high sampling costs associated with intractable normalization. ED also carries theoretical guarantees of higher training efficiency and reduced energy landscape distortion. Empirical evaluations on eight TAG datasets demonstrate that ERAlign obtains state-of-the-art performance across varying levels of supervision and cross-task transfer scenario.

Deep Learning · Graph Neural Networks

Lutong Wu, Shiying Cheng, Zhiqiang Wang, Jianqing Liang, Peng Song, Xizhao Luo, Jiye Liang

Explainability for graph neural networks (GNNs) aims to unveil the complex decision logic of learned models by identifying the most influential structures in the input graph, thereby improving transparency and trustworthiness. Existing post-hoc explainers typically extract a sparse key subgraph at a single scale as the explanation. However, a single-scale view often fails to capture multi-level semantics, and the optimization procedure may degenerate into a local search that is sensitive to initialization and noise, leading to unstable explanations and limited robustness. To address these issues, we propose MSExplainer, a multi-scale explainer for GNNs. MSExplainer couples multi-scale subgraph consistency guidance with single-scale adaptive subgraph learning under a parameter-sharing design. It simultaneously extracts multi-scale key subgraphs and complementary subgraphs, yielding a hierarchical decomposition of the original graph that covers semantics at different granularities and improves the robustness of subgraph extraction. Experiments on six benchmark datasets show that MSExplainer consistently outperforms prior methods in explanation accuracy and fidelity. Moreover, we theoretically prove the upper bound advantage of the multi-scale strategy in representation consistency, and derive that it achieves the same-order computational complexity as single-scale methods under the parameter-sharing mechanism, thus ensuring the high fidelity of key subgraphs while maintaining computational efficiency.

Applications · Neuroscience, Cognitive Science

Wentao Wang, Keren Gao, Guozhang Chen

Efficiently training recurrent neural networks to capture temporal structure is fundamental to machine intelligence, yet scaling gradient-based learning remains an open challenge. The standard global paradigm, backpropagation through time (BPTT), suffers from memory costs that scale linearly with sequence length and susceptibility to gradient instability. Conversely, biologically inspired local learning rules, while efficient, often introduce severe bias, failing to match the performance of global optimization. Inspired by the criticality observed in biological neural circuits, we introduce Criticality-driven Online Local Alignment (COLA). By leveraging the long-range spatiotemporal correlations inherent to the critical regime, COLA enables a strictly local learning rule to approximate the efficacy of global error propagation, thereby combining the advantages of online plasticity with the precision of gradient descent. Theoretically, for an RNN with $H$ hidden units, the method reduces learning complexity to a $O(H)$ auxiliary state with constant activation memory, independent of sequence length. Empirically, COLA matches BPTT on standard benchmarks and demonstrates superior robustness on stability-sensitive tasks. We support these results with a rigorous analysis on the approximation error, providing a theoretical foundation for reliable, scalable online learning.

Deep Learning · Graph Neural Networks

Fan Li, Xiaoyang Wang, Chen Chen, Wenjie Zhang

The increasing prevalence of large-scale hypergraphs poses significant computational challenges for hypergraph neural network (HNN) training. To address this, hypergraph condensation (HGC) distills large real hypergraphs into compact yet informative synthetic ones, going beyond traditional graph condensation (GC) methods limited to modeling pairwise relations. However, existing HGC methods rely on decoupled training architectures, where structure generators are pre-trained on the original hypergraph but not jointly optimized with condensed features during refinement, resulting in misaligned structures that degrade downstream utility. Moreover, trajectory-based optimization incurs substantial computational overhead in refinement, limiting condensation efficiency. To tackle these issues, we propose \textbf{A}nchor-guided \textbf{H}yper\textbf{G}raph \textbf{C}ondensation with \textbf{D}ual-level \textbf{D}iscrimination (\textbf{AHGCDD}), which consists of three key components: (1) a node initialization module based on Heat Kernel PageRank (HKPR) to encode structural knowledge into feature semantics; (2) an anchor-guided hyperedge synthesis scheme based on feature-level association for joint optimization of condensed features and structure; (3) a theoretically grounded dual-level discrimination objective for utility-preserving condensation without redundant HNN training. Extensive experiments demonstrate the superior effectiveness and efficiency of AHGCDD.

Deep Learning · Generative Models and Autoencoders

Alessandro Micheli, Yueqi Cao, Anthea Monod, Samir Bhatt

Computational optimal transport (OT) offers a principled framework for generative modeling. Neural OT methods, which use neural networks to learn an OT map (or potential) from data in an amortized way, can be evaluated out of sample after training, but existing approaches are tailored to Euclidean geometry. Extending neural OT to high-dimensional Riemannian manifolds remains an open challenge. In this paper, we prove that any method for OT on manifolds that produces discrete approximations of transport maps necessarily suffers from the curse of dimensionality: achieving a fixed accuracy requires a number of parameters that grows exponentially with the manifold dimension. Motivated by this limitation, we introduce Riemannian Neural OT (RNOT) maps, which are continuous neural-network parameterizations of OT maps on manifolds that avoid discretization and incorporate geometric structure by construction. Under mild regularity assumptions, we prove that RNOT maps approximate Riemannian OT maps with sub-exponential complexity in the dimension. Experiments on synthetic and real datasets demonstrate improved scalability and competitive performance relative to discretization-based baselines.