论文检索

Deep Learning · Large Language Models

Qianhao Yuan, Yanjiang Liu, Guozhao Mo, Yaojie Lu, Hongyu Lin, Jia Zheng, Ben He, Xianpei Han, Le Sun

Multimodal Large Language Models (MLLMs) mainly fall into two architectures, each involving a trade-off between training and inference efficiency: embedding space alignment (e.g. LLaVA series) is inefficient during inference, while cross-attention space alignment (e.g. Flamingo) is inefficient in training. A primary difference between them lies in whether each visual token attends to other tokens within the LLM backbones. To investigate whether this form of attention is essential for MLLMs, we propose NAEViT (No AttEntion from Visual Tokens), an attention mechanism that eliminates such interactions. Our pilot experiment shows that attention from visual tokens is highly redundant. Then, we introduce SAISA (Self-Attention Input Space Alignment), a novel architecture that enhances both training and inference efficiency. SAISA directly aligns visual features with the input spaces of NAEViT attention blocks, reducing computational overhead in both attention and FFNs. We conduct experiments on various baseline models, model sizes and training datasets. SAISA achieves superior performance compared to the baselines, while significantly reducing computational costs. Further ablation studies validate the effectiveness of SAISA across various LLMs and visual encoders

Theory · Domain Adaptation and Transfer Learning

Sagar Shrestha, Subash Timilsina, Hoang-Son Nguyen, Xiao Fu

Domain transfer (DT) maps source to target distributions and supports tasks such as unsupervised image-to-image translation, single-cell analysis, and cross-platform medical imaging. However, DT is fundamentally ill-posed: push-forward mappings are generally non-identifiable, as measure-preserving automorphisms (MPAs) preserve marginals while altering cross-domain correspondences, leading to content-misaligned translation. Recent work shows that MPAs can be eliminated by jointly transferring multiple corresponding source/target conditional distributions, but supervision signals labeling such conditionals are not always available in practice. We develop an alternative route to DT identifiability. Under a structural sparsity condition on the Jacobian support pattern, we show that distribution matching together with a single paired anchor sample suffices to identify the ground-truth transfer---requiring substantially less supervision than prior approaches. To enable practical high-dimensional learning, we further propose an efficient Jacobian sparsity regularizer based on randomized masked finite differences, yielding a scalable surrogate without explicit Jacobian evaluation. Empirical results on synthetic and real-world DT tasks validate the theory.

Applications · Neuroscience, Cognitive Science

Kartikay Agrawal, Vaishnavi N, Abhijeet Vikram, Vedant Sharma, Ayon Borthakur

Spiking neural networks have attracted increasing attention for their energy efficiency, multiplication-free computation, and sparse event-based processing. In parallel, state space models have emerged as a scalable alternative to transformers for long-range sequence modelling by avoiding quadratic dependence on sequence length. We propose here a spiking heterogeneous harmonic resonate-and-fire state space model (S$H^2$RFSSM), a second-order spiking SSM for classification and regression on ultra-long sequences. S$H^2$RFSSM outperforms transformers and first-order SSMs on average while eliminating matrix multiplications, making it highly suitable for resource-constrained applications. Furthermore, we introduce a kernel-based spiking regressor that enables accurate modelling of dependencies in sequences of up to 50k steps. We also observe a reduction in spiking operations and improved performance with heterogeneity and discretisation in harmonic resonate-and-fire neuronal layers. Overall, we evaluate Harmonic Resonate-and Fire layers across 17 diverse datasets, spanning sensors, time series, and classification to long-term forecasting. Our results demonstrate that S$H^2$RFSSM achieves superior long-range modelling capability with energy efficiency, positioning it as a strong candidate for signal processing on resource-constrained devices for human activity recognition, time series classification, and regression.

Deep Learning · Foundation Models

Mingqiao Ye, Zhaochong An, Zhitong Gao, Xian Liu, Oğuzhan Fatih Kar, Jesse Allardice, Roman Bachmann, David Mizrahi, François Fleuret, Chuan Li 等

Any-to-any modeling aims to flexibly relate arbitrary modalities within a single system, a requirement that arises across multimodal learning and scientific domains such as ecology and astronomy. However, existing any-to-any approaches are typically trained from scratch using encoder–decoder or diffusion architectures, limiting empirical performance and the use of pretrained models. We investigate decoder-only any-to-any multimodal modeling, which treats all modalities symmetrically and supports arbitrary modalities as inputs and outputs without modality-specific heads, losses, or task pipelines. As a consequence of this unified design, the resulting model MODUS naturally enables chained generation through intermediate modalities, cross-modal consistency verification, and analysis of visual representations by combining semantic and reconstruction features. Across a range of benchmarks, MODUS demonstrates strong out-of-the-box performance and flexible multimodal composition within a single model.

Applications · Chemistry, Physics, and Earth Sciences

Lingshi MENG, Haosen Shi, Sinno Jialin Pan

Physics-Informed Neural Networks (PINNs) represent a significant advancement in computational methods for solving partial differential equations (PDEs). However, the adoption of deeper neural network architectures presents significant challenges, as they struggle to address differential-related complications that arise during the computation of derivatives over the input of PINNs. These complications extend beyond traditional vanishing and exploding gradients to include vanishing and exploding differentials, with both phenomena becoming more severe as networks grow deeper. By examining the computation graph of derivatives in deep neural networks, we identify key bottlenecks causing numerical instabilities in deep architectures. In response, we introduce a novel approach that utilizes Coupling Layers with carefully regulated spectral norms of Jacobian matrices to stabilize and facilitate deep PINN training, effectively addressing differential-related challenges and improving model stability. Our proposed architecture successfully mitigates the fundamental constraints of deeper PINNs while maximizing their capabilities through consistent differential propagation. Comprehensive evaluations show that our approach surpasses conventional shallow PINN methods and alternative deep PINN designs across a range of challenging problems, particularly in cases featuring high-frequency solution components.

General Machine Learning · Transfer, Multitask and Meta-learning

Qi Tao, Jiarong Wen, Jing Yang, Guanlin Wu, Zhang Kaiyu, Yiqin Lv, Wumei Du, Xingxing Liang, Qi Wang

Importance-Weighted Neural Processes (IWNPs) provide a principled framework for probabilistic meta-learning by using multi-particle latent representations to approximate the marginal log-likelihood of task data tightly. However, this work reveals that the standard optimization of IWNPs suffers from the Matthew effect in the latent space, where high-likelihood particles dominate gradient signals. The neglect of lower-likelihood regions leads to poor tail-risk generation and unstable fast adaptation. While robust objectives such as $\text{CVaR}_\alpha$ can mitigate these risks, they often entail a trade-off that degrades average-case performance. This work proposes \underline{O}rder-\underline{S}tatistics Aligned \underline{N}eural \underline{P}rocesses (OS-NPs) to achieve latent space robust optimization without sacrificing average result. Specifically, we stratify multiple inference particles into disjoint difficulty bins based on order statistics and derive the regularized worst-case optimization framework for OS-NPs. Our method aligns the reduction of stratified order-statistic losses in IWNPs and provides a computationally efficient pipeline to implement. Extensive experiments demonstrate that the OS-NP constitutes stable, reliable probabilistic meta-learning that significantly enhances tail-risk robustness while maintaining or even improving average performance.

Applications · Computer Vision

Ming Dai, Sen Yang, Boqiang Duan, Boyuan Tong, Jiedong Zhuang, Wankou Yang, Jingdong Wang

Reasoning Video Object Segmentation (RVOS) demands a sophisticated integration of temporal dynamics, spatial details, and linguistic reasoning to achieve precise pixel-level localization. Existing methods are limited to reasoning over fixed initial inputs and lack the capacity to actively acquire further visual evidence, which is often essential for resolving complex references in long or intricate videos. To address this, we propose $\textbf{VideoSEG-O3}$, the first multi-turn reinforcement learning framework for RVOS that emulates the human $\textit{``coarse-to-fine''}$ cognitive process. It employs a $\textit{multi-turn temporal-spatial chain-of-thought}$ to capture fine-grained details by iteratively pinpointing critical intervals and keyframes. Additionally, to enable the policy to perceive segmentation quality beyond mere text probability of $\texttt{[SEG]}$ during the RL stage, we introduce $\textit{SEG-aware logit calibration}$, which integrates pixel-wise segmentation feedback directly into the token-level logits. Furthermore, we design a $\textit{decoupled thinking trace}$ to hierarchically decompose the reasoning process into temporal, spatial, and linguistic dimensions, and construct $\textbf{VTS-CoT}$, a specialized cold-start dataset featuring comprehensive reasoning trajectories. Extensive experiments demonstrate that VideoSEG-O3 achieves advanced performance across 8 mainstream RVOS benchmarks, particularly excelling in long-horizon and complex reasoning tasks.

Applications · Chemistry, Physics, and Earth Sciences

Miruna Cretu, John Bradshaw, Patricia Suriana, Saeed Saremi, Omar Mahmood, Kirill Shmilovich, Kangway Chuang, Vishnu Sresht, Colin A. Grambow

We present SynLaD, a latent diffusion framework for small-molecule generation that unifies 3D design objectives (what to make) with synthetic accessibility (how to make it). Current models typically optimize one objective at the expense of the other, creating a bottleneck for discovering high-scoring and experimentally testable molecules. SynLaD combines reaction-constrained generation with pharmacophore-conditioned 3D design by learning a latent space that decodes to both 3D structures and synthesis pathways. An encoder maps molecules to a latent representation used by two decoder heads: (i) a geometric head that reconstructs atom types and coordinates and (ii) an autoregressive synthesis head that outputs synthetic routes in a serialized, reaction-based notation. A diffusion transformer generates novel latents in the learned space, conditioned on pharmacophore profiles. Across analogue-generation tasks for bioactive ligands, SynLaD outperforms existing baselines in synthesizable and diverse hit generation, demonstrating that a single model can produce shape-accurate molecules with feasible synthesis plans.

Deep Learning · Generative Models and Autoencoders

Zhitong Gao, Parham Rezaei, Ali Cy, Mingqiao Ye, Nataša Jovanović, Jesse Allardice, Afshin Dehghan, Roman Bachmann, Oğuzhan Fatih Kar, Amir Zamir

Tokenization is a key component of autoregressive generative models, converting raw data into more manageable units for modeling. Commonly, tokens describe local information, such as regions of pixels in images or word pieces in text, and autoregressive generation commonly predicts these tokens in a fixed order. A worthwhile question is whether token structures affect the ability to steer the generation through test-time search, where multiple candidate generations are explored and evaluated by a verifier. Using image generation as our testbed, we hypothesize that recent 1D ordered tokenizers with coarse-to-fine structure can be more amenable to search than classical 2D grid structures. This is rooted in the fact that the intermediate states in coarse-to-fine sequences carry semantic meaning that verifiers can reliably evaluate, enabling effective steering during generation. We find that autoregressive models trained on coarse-to-fine ordered tokens exhibit improved test-time scaling behavior compared to grid-based counterparts. Moreover, we demonstrate that, thanks to the ordered structure, pure test-time search over token sequences (i.e., without training an autoregressive model) can perform training-free text-to-image generation when guided by an image-text verifier. Beyond this, we also systematically study how classical search algorithms (best-of-N, beam search, lookahead search) interact with different token structures, as well as the role of different verifiers and autoregressive priors in guiding the generation.

General Machine Learning · Sequential, Network, and Time Series Modeling

Haruka Ezoe, Hiroki Matsumoto, Ryohei Hisano

Dynamic relational data arise in many machine learning applications, yet their evolving structure poses challenges for learning representations that remain consistent and interpretable over time. A common approach is to learn time varying node embeddings, whose usefulness depends on well defined stability properties across nodes and across time. We introduce Unfolded Laplacian Spectral Embedding (ULSE), a principled extension of unfolded adjacency spectral embedding to normalized Laplacian operators, a setting where stability guarantees have remained out of reach. We prove that ULSE satisfies both cross-sectional and longitudinal stability under a dynamic stochastic block model. Moreover, the Laplacian formulation yields a dynamic Cheeger-type inequality linking the spectrum of the unfolded normalized Laplacian to worst case conductance over time, providing structural insight into the embeddings. Empirical results on synthetic and real world dynamic networks validate the theory.

General Machine Learning · Online Learning, Active Learning and Bandits

Amith Bhat Hosadurga Anand, Aadirupa Saha, Haipeng Luo

In this paper, we address the standard $K$-armed multi-armed bandit (MAB) with $M$ heterogeneous data sources, each exhibiting unknown and distinct noise variances, $\sigma_j^2$. We propose SOAR (Source-Optimistic Adaptive Regret Minimization), a novel algorithm that adaptively balances exploration and exploitation by jointly constructing upper confidence bounds for arm rewards and lower confidence bounds for data source variances. Our theoretical analysis establishes that SOAR achieves a regret bound of $\tilde{O}\left({\sigma^*}^2 \sum_{i=2}^K \tfrac{1}{\Delta_i}\right),$ along with a preprocessing cost that depends only on the problem parameters $\\{\sigma_j\\}_{j = 1}^M$, $K$, and grows at most logarithmically with the horizon $T$; where ${\sigma^\*}^2$ is the minimum source variance, and $\Delta_i$ denotes the suboptimality-gap of the $i$-th arm reward. The $\tilde O(.)$ notation hides the polylogarithmic factors in these problem parameters. This near-optimal instance dependence regret analysis of SOAR underscores its effectiveness in dynamically managing heteroscedastic noise without incurring significant overhead. Experiments on synthetic problem instances and a real dataset (MovieLens 25M) demonstrate that our method significantly outperforms baseline bandit algorithms in terms of regret performance. Our work opens a new direction for adaptively leveraging multiple heterogeneous data sources, extending beyond traditional bandit frameworks.

Theory · Domain Adaptation and Transfer Learning

Subash Timilsina, Hoang-Son Nguyen, Sagar Shrestha, Xiao Fu

Generative analysis often models multi-domain observations as nonlinear mixtures of domain-invariant content variables and domain-specific style variables. Identifying both factors from unpaired domains enables tasks such as domain transfer and counterfactual data generation. Prior work establishes identifiability under (block-wise) statistical independence between content and style, or via sparse Jacobian assumptions on the nonlinear mixing function, but such conditions can be restrictive and may not hold in practice. In this work, we introduce differential independence, a weaker structural condition requiring that infinitesimal variations in content and style induce orthogonal directions on the data manifold, thereby enabling identifiability even when content and style are dependent and the Jacobian is dense. We operationalize this condition through a blockwise orthogonality constraint on the Jacobian subspaces associated with content and style. To support high-dimensional generative models, we design a stochastic regularizer based on numerical Jacobian approximation, enabling scalable training in settings such as high-resolution image generation. Experiments across multiple datasets corroborate the identifiability analysis and demonstrate practical benefits on counterfactual generation and domain translation tasks.

Deep Learning · Generative Models and Autoencoders

Tongda Xu, Wendi Zheng, Jiajun He, Jose Miguel Hernandez-Lobato, Yan Wang, Ya-Qin Zhang, Jie Tang

Vector-quantized variational autoencoders (VQ-VAEs) are discrete autoencoders that compress images into discrete tokens. However, they are difficult to train due to discretization. In this paper, we propose a simple yet effective technique dubbed __Gaussian Quant (GQ)__, which first trains a Gaussian VAE under certain constraints and then converts it into a VQ-VAE without additional training. For conversion, GQ generates random Gaussian noise as a codebook and finds the closest noise vector to the posterior mean. Theoretically, we prove that when the logarithm of the codebook size exceeds the bits-back coding rate of the Gaussian VAE, a small quantization error is guaranteed. Practically, we propose a heuristic to train Gaussian VAEs for effective conversion, named the target divergence constraint (TDC). Empirically, we show that GQ outperforms previous VQ-VAEs, such as VQGAN, FSQ, LFQ, and BSQ, on both UNet and ViT architectures. Furthermore, TDC also improves previous Gaussian VAE discretization methods, such as TokenBridge. The source code is provided in the supplementary materials.

General Machine Learning · Data

Guanjie Zheng, Ziyang Su, Yiheng Wang, Yuhang Luo, Hongwei Zhang, Xuanhe Zhou, Linghe Kong, Fan Wu, Wen Ling

Road network data provides rich information about cities, but processing a large volume of worldwide OpenStreetMap (OSM) data is computationally intensive, and the resulting graphs are often difficult to unify for benchmarking downstream tasks. Existing graph learning benchmarks fail to capture the billion-scale and unique topological properties of real-world road networks, leading to a gap in our understanding of model scalability. To study and close this gap, we process OpenStreetMap data with distributed cloud computing using 5,000 cores and release OSM+, a structured worldwide 1-billion-vertex road network graph dataset designed for high accessibility and usability. OSM+ is open source and globally downloadable, and it provides an open-box graph structure together with an easy spatial query interface. We demonstrate the utility of OSM+ through three illustrative use cases: city boundary detection, traffic prediction, and traffic policy control. For traffic prediction, we construct a new 31-city benchmark by processing traffic data and combining it with OSM+, enabling broader spatial coverage and more comprehensive evaluation than previously frequently-used datasets, while scaling from hundreds of road network intersections to thousands. For traffic policy control, we release a new six-city dataset at a much larger scale, introducing challenges for thousand-scale multi-agent coordination. In addition, we provide comprehensive data processing tools that support integrating multimodal spatial-temporal data with OSM+ for geospatial foundation model training, thereby expediting the discovery of compelling scientific insights.

Reinforcement Learning · Online

Zhangyi Liu, Huaizhi Qu, Xiaowei Yin, He Sun, Yanjun Han, Tianlong Chen, Xinyu Yang

Test-time scaling can improve model performance by aggregating stochastic reasoning trajectories. However, achieving sample-efficient test-time self-consistency under a limited budget remains an open challenge. We introduce PETS (\textbf{P}rincipled and \textbf{E}fficient \textbf{T}est-Time \textbf{S}elf-Consistency), which initiates a principled study of trajectory allocation through an optimization framework. Central to our approach is the \emph{self-consistency rate}, a new measure defined as agreement with the infinite-budget majority vote. This formulation makes sample-efficient test-time allocation theoretically grounded and amenable to rigorous analysis. We study both offline and online settings. In the offline regime, where all questions are known in advance, we connect trajectory allocation to crowdsourcing, a classic and well-developed area, by modeling reasoning traces as workers. This perspective allows us to leverage rich existing theory, yielding theoretical guarantees and an efficient majority-voting-based allocation algorithm. In the online streaming regime, where questions arrive sequentially and allocations must be made on the fly, we propose a novel method inspired by the offline framework. Our approach adapts budgets to question difficulty while preserving strong theoretical guarantees and computational efficiency. Experiments show that PETS consistently outperforms uniform allocation. On GPQA, PETS achieves perfect self-consistency in both settings while reducing the sampling budget by up to $75\\%$ (offline) and $55\\%$ (online) relative to uniform allocation.

Deep Learning · Attention Mechanisms

Weikang Meng, Yadan Luo, Liangyu Huo, Yingjian Li, Yaowei Wang, Xin Li, Zheng Zhang

Linear attention mitigates the quadratic complexity of softmax attention but suffers from a critical loss of expressiveness. We identify two primary causes: (1) The normalization operation cancels the query norm, which breaks the correlation between a query's norm and the spikiness (entropy) of the attention distribution as in softmax attention. (2) Standard techniques for enforcing non-negativity cause destructive information loss by nullifying valid inner-product interactions. To address these challenges, we introduce **NaLaFormer**, a novel linear attention mechanism built upon a norm$\times$direction (ND) decomposition of the query and key vectors. We leverage each component to solve a distinct problem: The *query norm* is injected into our kernel to create a query-norm-aware map that restores the attention distribution's spikiness. The *direction vectors* are processed by a geometric, cosine-based similarity metric that guarantees non-negativity while preserving the rich, fine-grained information of the inner product. We validate NaLaFormer through a comprehensive multi-modal evaluation, where it sets new state-of-the-art benchmarks for linear attention. Our model achieves up to a 7.5\% accuracy gain on ImageNet-1K and a 4.7\% mIoU improvement on ADE20K over comparable baselines. It demonstrates profound efficiency, reducing peak memory by a transformative 92.3\% in token-intensive super-resolution tasks (70K+ tokens). NaLaFormer's versatility is further confirmed as it surpasses strong baselines like Mamba on common-sense reasoning and sets a new state-of-the-art on the Long Range Arena (LRA) benchmark. Source code can be found in the supplementary materials.

Theory · Online Learning and Bandits

Shogo Iwazaki

We study an algorithm-independent, worst-case lower bound for the Gaussian process (GP) bandit problem in the frequentist setting, where the reward function is fixed and has a bounded norm in the known reproducing kernel Hilbert space (RKHS). Specifically, we focus on the squared exponential (SE) kernel, one of the most widely used kernel functions in GP bandits. One of the remaining open questions for this problem is the gap in the *dimension-dependent* logarithmic factors between upper and lower bounds. This paper partially resolves this open question under a hyperspherical input domain. We show that any algorithm suffers $\Omega(\sqrt{T (\ln T)^{d} (\ln \ln T)^{-d}})$ cumulative regret, where $T$ and $d$ represent the total number of steps and the dimension of the hyperspherical domain, respectively. Regarding the simple regret, we show that any algorithm requires $\Omega(\epsilon^{-2}(\ln \frac{1}{\epsilon})^d (\ln \ln \frac{1}{\epsilon})^{-d})$ time steps to find an $\epsilon$-optimal point. We also provide the improved $O((\ln T)^{d+1}(\ln \ln T)^{-d})$ upper bound on the maximum information gain for the SE kernel. Our results guarantee the optimality of the existing best algorithm up to *dimension-independent* logarithmic factors under a hyperspherical input domain.

General Machine Learning · Transfer, Multitask and Meta-learning

Filippo Rinaldi, Aniello Panariello, Giacomo Salici, Angelo Porrello, Simone Calderara

Adapting large pre-trained models to downstream tasks often produces task-specific parameter updates that are expensive to relearn for every model variant. While recent work has shown that such updates can be transferred between models with identical architectures, transferring them across models of different widths remains largely unexplored. In this work, we introduce Theseus, a training-free method for transporting task-specific updates across heterogeneous models. Rather than matching parameters directly, we characterize a task update by the functional effect it induces on intermediate representations. We formalize task-vector transport as a functional matching problem on observed activations and show that, after aligning representation spaces via orthogonal Procrustes analysis, it admits a stable closed-form solution that preserves the geometry of the update. We evaluate Theseus on vision and language models across different widths, showing consistent improvements over strong baselines without additional training or backpropagation. Our results show that task updates can be meaningfully transferred across architectures when task identity is defined functionally rather than parametrically.

Reinforcement Learning · Everything Else

Yiming Fei, Ziming Wang, Rui Yan, Huajin Tang

Bottleneck states, which connect distinct regions of the state space, provide a principled and interpretable basis for constructing temporal abstractions in Hierarchical Reinforcement Learning (HRL). However, existing bottleneck identification methods primarily rely on topological analysis of the state-transition graph, limiting their scalability to high-dimensional or continuous domains. To address this challenge, we introduce Value Power Strength (VPS), a value function-based metric inspired by the analogy between the Bellman equation and Kirchhoff’s current law, to quantify bottleneck property via the diffusion of reward in Markov Decision Processes (MDPs). VPS is estimated efficiently using value functions learned from random reward signals and captures reward diffusion bottlenecks in both discrete and continuous state spaces. Leveraging VPS, we design options that guide agents toward or away from bottleneck regions. Experimental results on classic tabular domains, MiniGrid, and Atari 2600 games demonstrate that the VPS-based framework discovers semantically meaningful subgoals and substantially improves exploration efficiency.

Deep Learning · Other Representation Learning

Heda Zuo, Junxian Wu, Fengjie Lu, Pei Chen, Lingyun Sun, Weitao You

Vector Quantization (VQ) has been widely used in visual and audio representation due to its effectiveness in compressing high-dimensional signals. However, existing VQ methods often rely on large and unstructured codebooks, which leads to inefficient code utilization and frequent codebook collapse. In this paper, we propose *IChing* Vector Quantization (IVQ), a lightweight and structured vector quantization framework inspired by *IChing*. IVQ introduces binary hierarchical composition and geometric symmetry relations into the codebook design, enabling a compact set of structured codes to represent a large number of configurations while maintaining high utilization without codebook collapse. We conduct systematic comparisons between IVQ and several VQ variants mainly focusing on audio representation. Experimental results show that IVQ achieves superior quality with significantly smaller codebooks and consistently higher utilization rates. Auxiliary experiments on visual reconstruction and cross-modal alignment further validate the universality and robustness of our structured representation.