... International Conference on Learning Representations最新文献_第2页

Converge Faster, Talk Less: Hessian-Informed Federated Zeroth-Order Optimization. 收敛更快，说话更少：黑森信息联邦零阶优化。

... International Conference on Learning Representations Pub Date : 2026-01-01

Zhe Li, Bicheng Ying, Zidong Liu, Chaosheng Dong, Haibo Yang

{"title":"Converge Faster, Talk Less: Hessian-Informed Federated Zeroth-Order Optimization.","authors":"Zhe Li, Bicheng Ying, Zidong Liu, Chaosheng Dong, Haibo Yang","doi":"","DOIUrl":"","url":null,"abstract":"Zeroth-order (ZO) optimization enables dimension-free communication in federated learning (FL), making it attractive for fine-tuning of large language models (LLMs) due to significant communication savings. However, existing ZO-FL methods largely overlook curvature information, despite its well-established benefits for convergence acceleration. To address this, we propose HiSo, a Hessian-informed ZO federated optimization method that accelerates convergence by leveraging global diagonal Hessian approximations, while strictly preserving scalar-only communication without transmitting any second-order information. Theoretically, for non-convex functions, we show that HiSo can achieve an accelerated convergence rate that is independent of the Lipschitz constant <math><mi>L</mi></math> and model dimension <math><mi>d</mi></math> under some Hessian approximation assumptions, offering a plausible explanation for the observed phenomenon of ZO convergence being much faster than its worst-case <math><mi>𝒪</mi> <mo>(</mo> <mi>d</mi> <mo>)</mo></math> -bound. Empirically, across diverse LLM fine-tuning benchmarks, HiSo delivers a 1~5× speedup in communication rounds over existing state-of-the-art ZO-FL baselines. This superior convergence not only cuts communication costs but also provides strong empirical evidence that Hessian information acts as an effective accelerator in federated ZO optimization settings.","PeriodicalId":521258,"journal":{"name":"... International Conference on Learning Representations","volume":"2026 ","pages":"37066-37099"},"PeriodicalIF":0.0,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13461244/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148715249","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

AVERE: Improving Audiovisual Emotion Reasoning with Preference Optimization. AVERE：用偏好优化改进视听情感推理。

... International Conference on Learning Representations Pub Date : 2026-01-01

Ashutosh Chaubey, Jiacheng Pang, Maksim Siniukov, Mohammad Soleymani

{"title":"AVERE: Improving Audiovisual Emotion Reasoning with Preference Optimization.","authors":"Ashutosh Chaubey, Jiacheng Pang, Maksim Siniukov, Mohammad Soleymani","doi":"","DOIUrl":"","url":null,"abstract":"Emotion understanding is essential for building socially intelligent agents. Although recent multimodal large language models (MLLMs) have shown strong performance on this task, two key challenges remain: (i) spurious associations between emotions and irrelevant audiovisual cues (reasoning errors) and (ii) hallucination of audiovisual cues (perception errors) driven by text priors in the language model backbone. To quantify and understand these issues, we introduce EmoReAlM, a benchmark designed to evaluate MLLMs for cue-emotion associations, hallucinations and modality agreement. We then propose AVEm-DPO, a preference optimization technique that aligns model responses with both audiovisual inputs and emotion-centric queries. Specifically, we construct preferences over (i) responses exhibiting spurious associations or hallucinations and (ii) audiovisual input pairs guided by textual prompts. We also include a regularization term that penalizes reliance on text priors, thereby mitigating modality-specific cue hallucinations. Experimental results on DFEW, RAVDESS and EMER demonstrate that our method significantly improves the performance of the reference baseline models (6-19% of relative performance) in zero-shot settings. By providing both a rigorous benchmark and a robust optimization framework, this work enables principled evaluation and improvement of MLLMs for emotion understanding and social AI. Code, models and benchmark at our project page - avere-iclr.github.io.","PeriodicalId":521258,"journal":{"name":"... International Conference on Learning Representations","volume":"2026 ","pages":"14110-14154"},"PeriodicalIF":0.0,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13615894/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149067299","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Unified Brain Surface and Volume Registration. 统一脑表面和脑容量登记。

... International Conference on Learning Representations Pub Date : 2026-01-01

S Mazdak Abulnaga, Andrew Hoopes, Malte Hoffmann, Robin Magnet, Maks Ovsjanikov, Lilla Zöllei, John Guttag, Bruce Fischl, Adrian V Dalca

{"title":"Unified Brain Surface and Volume Registration.","authors":"S Mazdak Abulnaga, Andrew Hoopes, Malte Hoffmann, Robin Magnet, Maks Ovsjanikov, Lilla Zöllei, John Guttag, Bruce Fischl, Adrian V Dalca","doi":"","DOIUrl":"","url":null,"abstract":"Accurate registration of brain MRI scans is fundamental for cross-subject analysis in neuroscientific studies. This involves aligning both the cortical surface of the brain and the interior volume. Traditional methods treat volumetric and surface-based registration separately, which often leads to inconsistencies that limit downstream analyses. We propose a deep learning framework, NeurAlign, that registers 3D brain MRI images by jointly aligning both cortical and subcortical regions through a unified volume-and-surface-based representation. Our approach leverages an intermediate spherical coordinate space to bridge anatomical surface topology with volumetric anatomy, enabling consistent and anatomically accurate alignment. By integrating spherical registration into the learning, our method ensures geometric coherence between volume and surface domains. In a series of experiments on both in-domain and out-of-domain datasets, our method consistently outperforms both classical and machine learning-based registration methods-improving the Dice score by up to 7 points while maintaining regular deformation fields. Additionally, it is orders of magnitude faster than the standard method for this task, and is simpler to use because it requires no additional inputs beyond an MRI scan. With its superior accuracy, fast inference, and ease of use, NeurAlign sets a new standard for joint cortical and subcortical registration.","PeriodicalId":521258,"journal":{"name":"... International Conference on Learning Representations","volume":"2026 ","pages":"132177-132193"},"PeriodicalIF":0.0,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13354064/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148428658","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

A TALE OF TWO TAILS: PREFERRED AND ANTI-PREFERRED NATURAL STIMULI IN VISUAL CORTEX. 两个尾巴的故事：视觉皮层中首选和反首选的自然刺激。

... International Conference on Learning Representations Pub Date : 2026-01-01

Rabia Gondur, Patricia L Stan, Matthew A Smith, Benjamin R Cowley

{"title":"A TALE OF TWO TAILS: PREFERRED AND ANTI-PREFERRED NATURAL STIMULI IN VISUAL CORTEX.","authors":"Rabia Gondur, Patricia L Stan, Matthew A Smith, Benjamin R Cowley","doi":"","DOIUrl":"","url":null,"abstract":"An ongoing quest in neuroscience is to find the preferred stimulus of a sensory neuron. This search lays the foundation for understanding how selectivity emerges in the primate visual stream-from simple edge-detecting neurons to highly-selective face neurons-as well as for the architectures and activation functions of deep neural networks. The prevailing notion is that a visual neuron primarily responds to a single preferred visual feature, like an oriented edge or the shape of an object, resulting in a \"one-tailed\" distribution of responses to natural images. However, surprisingly, we instead find \"two-tailed\" response distributions of primate visual cortical neurons, suggesting that these neurons have both preferred and anti-preferred stimuli. We experimentally validated anti-preferred stimuli by recording responses from macaque V4 to model-optimized stimuli. We find that these anti-preferred stimuli are important for describing a neuron's tuning, as both preferred and anti-preferred images are needed to predict a neuron's responses to natural images. Moreover, in a psychophysics task, humans rely on anti-preferred images to interpret and predict V4 stimulus tuning; this was not the case for internal units from a deep neural network. Interestingly, we find no discernible differences in image statistics between preferred and anti-preferred images. This suggests that by encoding anti-preferred features, a V4 population seemingly doubles its capacity for feature selectivity, allowing for a more flexible downstream readout. Overall, we establish anti-preferred stimuli as an important encoding property of V4 neurons. Our work embarks on a new quest in neuroscience to search for anti-preferred stimuli along the visual stream and offers a new perspective on how feature selectivity arises in the visual cortex and deep neural networks.","PeriodicalId":521258,"journal":{"name":"... International Conference on Learning Representations","volume":"2026 ","pages":"98177-98204"},"PeriodicalIF":0.0,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13440251/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148681639","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

ICYM²I: The illusion of multimodal informativeness under missingness. ICYM2I：缺失下的多模态信息错觉。

... International Conference on Learning Representations Pub Date : 2026-01-01

Young Sang Choi, Vincent Jeanselme, Pierre Elias, Shalmali Joshi

{"title":"ICYM2I: The illusion of multimodal informativeness under missingness.","authors":"Young Sang Choi, Vincent Jeanselme, Pierre Elias, Shalmali Joshi","doi":"","DOIUrl":"","url":null,"abstract":"Multimodal learning is of continued interest in artificial intelligence-based applications, motivated by the potential information gain from combining different data modalities. However, modalities observed in the source environment may differ from the modalities observed in the target environment due to multiple factors, including cost, hardware failure, or the perceived informativeness of a given modality. This change in missingness patterns between the source and target environment has not been carefully studied. Naïve estimation of the information gain associated with including an additional modality without accounting for missingness may result in improper estimates of that modality's value in the target environment. We formalize the problem of missingness, demonstrate its ubiquity, and show that the subsequent distribution shift induces bias when the missingness process is not explicitly accounted for. To address this issue, we introduce ICYM2I (In Case You Multimodal Missed It), a framework for the evaluation of predictive performance and information gain under missingness through inverse probability weighting-based correction. We demonstrate the importance of the proposed adjustment to estimate information gain under missingness on synthetic, semi-synthetic, and real-world datasets.","PeriodicalId":521258,"journal":{"name":"... International Conference on Learning Representations","volume":"2026 ","pages":"84620-84648"},"PeriodicalIF":0.0,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13460310/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148715258","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Bi-Lipschitz Autoencoder With Injectivity Guarantee. 具有注入性保证的Bi-Lipschitz自编码器。

... International Conference on Learning Representations Pub Date : 2026-01-01

Qipeng Zhan, Zhuoping Zhou, Zexuan Wang, Qi Long, Li Shen

{"title":"Bi-Lipschitz Autoencoder With Injectivity Guarantee.","authors":"Qipeng Zhan, Zhuoping Zhou, Zexuan Wang, Qi Long, Li Shen","doi":"","DOIUrl":"","url":null,"abstract":"Autoencoders are widely used for dimensionality reduction, based on the assumption that high-dimensional data lies on low-dimensional manifolds. Regularized autoencoders aim to preserve manifold geometry during dimensionality reduction, but existing approaches often suffer from non-injective mappings and overly rigid constraints that limit their effectiveness and robustness. In this work, we identify encoder non-injectivity as a core bottleneck that leads to poor convergence and distorted latent representations. To ensure robustness across data distributions, we formalize the concept of admissible regularization and provide sufficient conditions for its satisfaction. In this work, we propose the Bi-Lipschitz Autoencoder (BLAE), which introduces two key innovations: (1) an injective regularization scheme based on a separation criterion to eliminate pathological local minima, and (2) a bi-Lipschitz relaxation that preserves geometry and exhibits robustness to data distribution drift. Empirical results on diverse datasets show that BLAE consistently outperforms existing methods in preserving manifold structure while remaining resilient to sampling sparsity and distribution shifts. Code is available at https://github.com/qipengz/BLAE.","PeriodicalId":521258,"journal":{"name":"... International Conference on Learning Representations","volume":"2026 ","pages":"118572-118598"},"PeriodicalIF":0.0,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13532366/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148876573","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Improving the Trade-off Between Watermark Strength and Speculative Sampling Efficiency for Language Models. 改进语言模型中水印强度和推测采样效率之间的权衡。

... International Conference on Learning Representations Pub Date : 2026-01-01

Weiqing He, Xiang Li, Li Shen, Weijie Su, Qi Long

{"title":"Improving the Trade-off Between Watermark Strength and Speculative Sampling Efficiency for Language Models.","authors":"Weiqing He, Xiang Li, Li Shen, Weijie Su, Qi Long","doi":"","DOIUrl":"","url":null,"abstract":"Watermarking is a principled approach for tracing the provenance of large language model (LLM) outputs, but its deployment in practice is hindered by inference inefficiency. Speculative sampling accelerates inference, with efficiency improving as the acceptance rate between draft and target models increases. Yet recent work reveals a fundamental trade-off: higher watermark strength reduces acceptance, preventing their simultaneous achievement. We revisit this trade-off and show it is not absolute. We introduce a quantitative measure of watermark strength that governs statistical detectability and is maximized when tokens are deterministic functions of pseudorandom numbers. Using this measure, we fully characterize the trade-off as a constrained optimization problem and derive explicit Pareto curves for two existing watermarking schemes. Finally, we introduce a principled mechanism that injects pseudorandomness into draft-token acceptance, ensuring maximal watermark strength while maintaining speculative sampling efficiency. Experiments further show that this approach improves detectability without sacrificing efficiency. Our findings uncover a principle that unites speculative sampling and watermarking, paving the way for their efficient and practical deployment.","PeriodicalId":521258,"journal":{"name":"... International Conference on Learning Representations","volume":"2026 ","pages":"30472-30496"},"PeriodicalIF":0.0,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13548028/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148908666","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Beyond Accuracy: Are Time Series Foundation Models Well-Calibrated? 超越准确性：时间序列基础模型是否校准良好？

... International Conference on Learning Representations Pub Date : 2026-01-01

Coen Adler, Yuxin Chang, Felix Draxler, Samar Abdi, Padhraic Smyth

引用次数: 0

Disentangled Representation Learning for Parametric Partial Differential Equations. 参数偏微分方程的解纠缠表示学习。

... International Conference on Learning Representations Pub Date : 2026-01-01

Ning Liu, Lu Zhang, Tian Gao, Yue Yu

{"title":"Disentangled Representation Learning for Parametric Partial Differential Equations.","authors":"Ning Liu, Lu Zhang, Tian Gao, Yue Yu","doi":"","DOIUrl":"","url":null,"abstract":"Neural operators (NOs) excel at learning mappings between function spaces, serving as efficient forward solution approximators for PDE-governed systems. However, as black-box solvers, they offer limited insight into the underlying physical mechanism, due to the lack of interpretable representations of the physical parameters that drive the system. To tackle this challenge, we propose a new paradigm for learning disentangled representations from NO parameters, thereby effectively solving an inverse problem. Specifically, we introduce DisentangO, a novel hyper-neural operator architecture designed to unveil and disentangle latent physical factors of variation embedded within the black-box neural operator parameters. At the core of DisentangO is a multi-task NO architecture that distills the varying parameters of the governing PDE through a task-wise adaptive layer, alongside a variational autoencoder that disentangles these variations into identifiable latent factors. By learning these disentangled representations, DisentangO not only enhances physical interpretability but also enables more robust generalization across diverse systems. Empirical evaluations across supervised, semi-supervised, and unsupervised learning contexts show that DisentangO effectively extracts meaningful and interpretable latent features, bridging the gap between predictive performance and physical understanding in neural operator frameworks. Our code and data accompanying this paper are available at https://github.com/ningliu-iga/DisentangO.","PeriodicalId":521258,"journal":{"name":"... International Conference on Learning Representations","volume":"2026 ","pages":"147920-147949"},"PeriodicalIF":0.0,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13460308/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148715255","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Diffusion-DFL: Decision-focused Diffusion Models for Stochastic Optimization. 扩散- dfl：基于决策的随机优化扩散模型。

... International Conference on Learning Representations Pub Date : 2026-01-01

Zihao Zhao, Christopher Yeh, Lingkai Kong, Kai Wang

{"title":"Diffusion-DFL: Decision-focused Diffusion Models for Stochastic Optimization.","authors":"Zihao Zhao, Christopher Yeh, Lingkai Kong, Kai Wang","doi":"","DOIUrl":"","url":null,"abstract":"Decision-focused learning (DFL) integrates predictive modeling and optimization by training predictors to optimize the downstream decision target rather than merely minimizing prediction error. To date, existing DFL methods typically rely on deterministic point predictions, which are often insufficient to capture the intrinsic stochasticity of real-world environments. To address this challenge, we propose the first diffusion-based DFL approach, which trains a diffusion model to represent the distribution of uncertain parameters and optimizes the decision by solving a stochastic optimization with samples drawn from the diffusion model. Our contributions are twofold. First, we formulate diffusion DFL using the reparameterization trick, enabling end-to-end training through diffusion. While effective, it is memory and compute-intensive due to the need to differentiate through the diffusion sampling process. Second, we propose a lightweight score function estimator that uses only several forward diffusion passes and avoids backpropagation through the sampling. This follows from our results that backpropagating through stochastic optimization can be approximated by a weighted score function formulation. We empirically show that our diffusion DFL approach consistently outperforms strong baselines in decision quality. The source code for all experiments is available at https://github.com/GT-KOALA/Diffusion_DFL.","PeriodicalId":521258,"journal":{"name":"... International Conference on Learning Representations","volume":"2026 ","pages":"36834-36859"},"PeriodicalIF":0.0,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13460018/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148715221","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0