CAMEx: Curvature-aware Merging of Experts

• URL: http://arxiv.org/abs/2502.18821v2
• Date: Mon, 03 Mar 2025 16:12:50 GMT
• Title: CAMEx: Curvature-aware Merging of Experts
• Authors: Dung V. Nguyen, Minh H. Nguyen, Luc Q. Nguyen, Rachel S. Y. Teo, Tan M. Nguyen, Linh Duy Tran,
• Abstract summary: Existing methods for merging experts during model training and fine-tuning rely on Euclidean geometry.<n>Curvature-aware merging methods require additional information and computational resources to approximate the Fisher Information Matrix.<n>We introduce CAMEx, a novel expert merging protocol that incorporates natural gradients to account for the non-Euclidean curvature of the parameter manifold.
• Score: 1.5479848902142663
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Existing methods for merging experts during model training and fine-tuning predominantly rely on Euclidean geometry, which assumes a flat parameter space. This assumption can limit the model's generalization ability, especially during the pre-training phase, where the parameter manifold might exhibit more complex curvature. Curvature-aware merging methods typically require additional information and computational resources to approximate the Fisher Information Matrix, adding memory overhead. In this paper, we introduce CAMEx (Curvature-Aware Merging of Experts), a novel expert merging protocol that incorporates natural gradients to account for the non-Euclidean curvature of the parameter manifold. By leveraging natural gradients, CAMEx adapts more effectively to the structure of the parameter space, improving alignment between model updates and the manifold's geometry. This approach enhances both pre-training and fine-tuning, resulting in better optimization trajectories and improved generalization without the substantial memory overhead typically associated with curvature-aware methods. Our contributions are threefold: (1) CAMEx significantly outperforms traditional Euclidean-based expert merging techniques across various natural language processing tasks, leading to enhanced performance during pre-training and fine-tuning; (2) we introduce a dynamic merging architecture that optimizes resource utilization, achieving high performance while reducing computational costs, facilitating efficient scaling of large language models; and (3) we provide both theoretical and empirical evidence to demonstrate the efficiency of our proposed method. The code is publicly available at: https://github.com/kpup1710/CAMEx.

Related papers

• Parameter-Efficient Fine-Tuning of LLMs with Mixture of Space Experts [20.82313207866023]
  We propose a unified framework that leverages multiple geometric spaces simultaneously to learn curvature-aware representations.<n>We develop MoSLoRA, which extends Low-Rank Adaptation (LoRA) with heterogeneous geometric experts.<n>Our experiments across diverse benchmarks demonstrate that MoSLoRA consistently outperforms strong baselines.
  arXiv  Detail & Related papers  (2026-02-16T06:07:32Z)
• Merging Beyond: Streaming LLM Updates via Activation-Guided Rotations [55.047454145941366]
  Streaming Merging is an innovative model updating paradigm that conceptualizes merging as an iterative optimization process.<n> ARM is a strategy designed to approximate gradient descent dynamics.<n> ARM requires only early SFT checkpoints and, through iterative merging, surpasses the fully converged SFT model.
  arXiv  Detail & Related papers  (2026-02-03T08:15:57Z)
• Learning Geometry: A Framework for Building Adaptive Manifold Models through Metric Optimization [8.201374511929538]
  This paper proposes a novel paradigm for machine learning that moves beyond traditional parameter optimization.<n>We optimize the metric tensor field on a manifold with a predefined topology, thereby dynamically shaping the geometric structure of the model space.<n>This work lays a solid foundation for constructing fully dynamic "meta-learners" capable of autonomously evolving their geometry and topology.
  arXiv  Detail & Related papers  (2025-10-30T01:53:32Z)
• Preconditioned Norms: A Unified Framework for Steepest Descent, Quasi-Newton and Adaptive Methods [50.070182958880146]
  We propose a unified framework generalizing descent, quasi-Newton methods, and adaptive methods through the novel notion of preconditioned matrix norms.<n>Within this framework, we provide the first systematic treatment of affine and scale invariance in the matrix- parameterized setting.<n>We introduce two new methods, $ttMuAdam$ and $texttMuAdam-SANIA$, which combine the spectral geometry of Muon with Adam-style preconditioning.
  arXiv  Detail & Related papers  (2025-10-12T19:39:41Z)
• MEPT: Mixture of Expert Prompt Tuning as a Manifold Mapper [75.6582687942241]
  We propose Mixture of Expert Prompt Tuning (MEPT) as an effective and efficient manifold-mapping framework.<n>MEPT integrates multiple prompt experts to adaptively learn diverse and non-stationary data distributions.<n> Empirical evaluations demonstrate that MEPT outperforms several state-of-the-art parameter efficient baselines on SuperGLUE.
  arXiv  Detail & Related papers  (2025-08-31T21:19:25Z)
• L-SR1: Learned Symmetric-Rank-One Preconditioning [5.421390145168128]
  End-to-end deep learning has achieved impressive results but remains limited by its reliance on large labeled datasets.<n>In contrast, classical optimization methods are data-efficient and lightweight but often suffer from slow convergence.<n>We propose a novel learned second-order vectors that introduces a trainable preconditioning unit to enhance the classical Symmetric-Rank-One algorithm.
  arXiv  Detail & Related papers  (2025-08-17T07:37:29Z)
• Geometric Operator Learning with Optimal Transport [77.16909146519227]
  We propose integrating optimal transport (OT) into operator learning for partial differential equations (PDEs) on complex geometries.<n>For 3D simulations focused on surfaces, our OT-based neural operator embeds the surface geometry into a 2D parameterized latent space.<n> Experiments with Reynolds-averaged Navier-Stokes equations (RANS) on the ShapeNet-Car and DrivAerNet-Car datasets show that our method achieves better accuracy and also reduces computational expenses.
  arXiv  Detail & Related papers  (2025-07-26T21:28:25Z)
• Curve-Aware Gaussian Splatting for 3D Parametric Curve Reconstruction [14.628742412460346]
  This paper presents an end-to-end framework for reconstructing 3D parametric curves directly from multi-view edge maps.<n>We propose a novel bi-directional coupling mechanism between parametric curves and edge-oriented Gaussian components.<n>Our method significantly reduces the parameter count during training, achieving both higher efficiency and superior performance compared to existing approaches.
  arXiv  Detail & Related papers  (2025-06-26T15:48:08Z)
• Curvature Enhanced Data Augmentation for Regression [4.910937238451485]
  We introduce the Curvature-Enhanced Manifold Sampling (CEMS) method for regression tasks.<n>CEMS delivers superior performance in both in-distribution and out-of-distribution scenarios.
  arXiv  Detail & Related papers  (2025-06-07T16:18:37Z)
• Generalized Tensor-based Parameter-Efficient Fine-Tuning via Lie Group Transformations [50.010924231754856]
  Adapting pre-trained foundation models for diverse downstream tasks is a core practice in artificial intelligence. To overcome this, parameter-efficient fine-tuning (PEFT) methods like LoRA have emerged and are becoming a growing research focus. We propose a generalization that extends matrix-based PEFT methods to higher-dimensional parameter spaces without compromising their structural properties.
  arXiv  Detail & Related papers  (2025-04-01T14:36:45Z)
• Riemannian Geometric-based Meta Learning [8.365106891566725]
  "Learning to learn" aims to enable models to quickly adapt to new tasks with minimal data. Traditional methods like Model-Agnostic Meta-Learning (MAML) often struggle to capture complex learning dynamics. We propose Stiefel-MAML, which integrates Riemannian geometry by optimizing within the Stiefel manifold.
  arXiv  Detail & Related papers  (2025-03-14T01:34:55Z)
• Training Deep Learning Models with Norm-Constrained LMOs [56.00317694850397]
  We study optimization methods that leverage the linear minimization oracle (LMO) over a norm-ball. We propose a new family of algorithms that uses the LMO to adapt to the geometry of the problem and, perhaps surprisingly, show that they can be applied to unconstrained problems.
  arXiv  Detail & Related papers  (2025-02-11T13:10:34Z)
• ALoRE: Efficient Visual Adaptation via Aggregating Low Rank Experts [71.91042186338163]
  ALoRE is a novel PETL method that reuses the hypercomplex parameterized space constructed by Kronecker product to Aggregate Low Rank Experts. Thanks to the artful design, ALoRE maintains negligible extra parameters and can be effortlessly merged into the frozen backbone.
  arXiv  Detail & Related papers  (2024-12-11T12:31:30Z)
• FORML: A Riemannian Hessian-free Method for Meta-learning on Stiefel Manifolds [4.757859522106933]
  This paper introduces a Hessian-free approach that uses a first-order approximation of derivatives on the Stiefel manifold. Our method significantly reduces the computational load and memory footprint.
  arXiv  Detail & Related papers  (2024-02-28T10:57:30Z)
• The Convex Landscape of Neural Networks: Characterizing Global Optima and Stationary Points via Lasso Models [75.33431791218302]
  Deep Neural Network Network (DNN) models are used for programming purposes. In this paper we examine the use of convex neural recovery models. We show that all the stationary non-dimensional objective objective can be characterized as the standard a global subsampled convex solvers program. We also show that all the stationary non-dimensional objective objective can be characterized as the standard a global subsampled convex solvers program.
  arXiv  Detail & Related papers  (2023-12-19T23:04:56Z)
• GloptiNets: Scalable Non-Convex Optimization with Certificates [61.50835040805378]
  We present a novel approach to non-cube optimization with certificates, which handles smooth functions on the hypercube or on the torus. By exploiting the regularity of the target function intrinsic in the decay of its spectrum, we allow at the same time to obtain precise certificates and leverage the advanced and powerful neural networks.
  arXiv  Detail & Related papers  (2023-06-26T09:42:59Z)
• Scaling Pre-trained Language Models to Deeper via Parameter-efficient Architecture [68.13678918660872]
  We design a more capable parameter-sharing architecture based on matrix product operator (MPO) MPO decomposition can reorganize and factorize the information of a parameter matrix into two parts. Our architecture shares the central tensor across all layers for reducing the model size.
  arXiv  Detail & Related papers  (2023-03-27T02:34:09Z)
• Learning Augmentation Distributions using Transformed Risk Minimization [47.236227685707526]
  We propose a new emphTransformed Risk Minimization (TRM) framework as an extension of classical risk minimization. As a key application, we focus on learning augmentations to improve classification performance with a given class of predictors.
  arXiv  Detail & Related papers  (2021-11-16T02:07:20Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.