Gating is Weighting: Understanding Gated Linear Attention through In-context Learning

• URL: http://arxiv.org/abs/2504.04308v1
• Date: Sun, 06 Apr 2025 00:37:36 GMT
• Title: Gating is Weighting: Understanding Gated Linear Attention through In-context Learning
• Authors: Yingcong Li, Davoud Ataee Tarzanagh, Ankit Singh Rawat, Maryam Fazel, Samet Oymak,
• Abstract summary: Gated Linear Attention (GLA) architectures include competitive models such as Mamba and RWKV.<n>We show that a multilayer GLA can implement a general class of Weighted Preconditioned Gradient Descent (WPGD) algorithms.<n>Under mild conditions, we establish the existence and uniqueness (up to scaling) of a global minimum, corresponding to a unique WPGD solution.
• Score: 48.90556054777393
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Linear attention methods offer a compelling alternative to softmax attention due to their efficiency in recurrent decoding. Recent research has focused on enhancing standard linear attention by incorporating gating while retaining its computational benefits. Such Gated Linear Attention (GLA) architectures include competitive models such as Mamba and RWKV. In this work, we investigate the in-context learning capabilities of the GLA model and make the following contributions. We show that a multilayer GLA can implement a general class of Weighted Preconditioned Gradient Descent (WPGD) algorithms with data-dependent weights. These weights are induced by the gating mechanism and the input, enabling the model to control the contribution of individual tokens to prediction. To further understand the mechanics of this weighting, we introduce a novel data model with multitask prompts and characterize the optimization landscape of learning a WPGD algorithm. Under mild conditions, we establish the existence and uniqueness (up to scaling) of a global minimum, corresponding to a unique WPGD solution. Finally, we translate these findings to explore the optimization landscape of GLA and shed light on how gating facilitates context-aware learning and when it is provably better than vanilla linear attention.

Related papers

• Improving Autoregressive Visual Generation with Cluster-Oriented Token Prediction [52.09472099976885]
  IAR is an Improved AutoRegressive Visual Generation Method that enhances the training efficiency and generation quality of LLM-based visual generation models.<n>Our method consistently enhances the model training efficiency and performance from 100M to 1.4B, reducing the training time by half while achieving the same FID.
  arXiv  Detail & Related papers  (2025-01-01T15:58:51Z)
• Simple Is Effective: The Roles of Graphs and Large Language Models in Knowledge-Graph-Based Retrieval-Augmented Generation [9.844598565914055]
  Large Language Models (LLMs) demonstrate strong reasoning abilities but face limitations such as hallucinations and outdated knowledge.<n>We introduce SubgraphRAG, extending the Knowledge Graph (KG)-based Retrieval-Augmented Generation (RAG) framework that retrieves subgraphs.<n>Our approach innovatively integrates a lightweight multilayer perceptron with a parallel triple-scoring mechanism for efficient and flexible subgraph retrieval.
  arXiv  Detail & Related papers  (2024-10-28T04:39:32Z)
• Bridging Large Language Models and Graph Structure Learning Models for Robust Representation Learning [22.993015048941444]
  Graph representation learning is crucial for real-world applications but often encounters pervasive noise. We introduce LangGSL, a framework that integrates the complementary strengths of pre-trained language models and graph structure learning models.
  arXiv  Detail & Related papers  (2024-10-15T22:43:32Z)
• How to Make LLMs Strong Node Classifiers? [70.14063765424012]
  Language Models (LMs) are challenging the dominance of domain-specific models, such as Graph Neural Networks (GNNs) and Graph Transformers (GTs)<n>We propose a novel approach that empowers off-the-shelf LMs to achieve performance comparable to state-of-the-art (SOTA) GNNs on node classification tasks.
  arXiv  Detail & Related papers  (2024-10-03T08:27:54Z)
• On the Generalization Capability of Temporal Graph Learning Algorithms: Theoretical Insights and a Simpler Method [59.52204415829695]
  Temporal Graph Learning (TGL) has become a prevalent technique across diverse real-world applications. This paper investigates the generalization ability of different TGL algorithms. We propose a simplified TGL network, which enjoys a small generalization error, improved overall performance, and lower model complexity.
  arXiv  Detail & Related papers  (2024-02-26T08:22:22Z)
• Joint Graph Learning and Model Fitting in Laplacian Regularized Stratified Models [5.933030735757292]
  Laplacian regularized stratified models (LRSM) are models that utilize the explicit or implicit network structure of the sub-problems. This paper shows the importance and sensitivity of graph weights in LRSM, and provably show that the sensitivity can be arbitrarily large. We propose a generic approach to jointly learn the graph while fitting the model parameters by solving a single optimization problem.
  arXiv  Detail & Related papers  (2023-05-04T06:06:29Z)
• Semi-Supervised Graph Learning Meets Dimensionality Reduction [0.0]
  Semi-supervised learning (SSL) has recently received increased attention from machine learning researchers. In this work, we investigate the use of dimensionality reduction techniques such as PCA, t-SNE, and UMAP to see their effect on the performance of graph neural networks (GNNs) Our benchmarks and clustering visualizations demonstrate that, under certain conditions, employing a priori and a posteriori dimensionality reduction to GNN inputs and outputs, respectively, can simultaneously improve the effectiveness of semi-supervised node label propagation and node clustering.
  arXiv  Detail & Related papers  (2022-03-23T16:31:53Z)
• Towards Unsupervised Deep Graph Structure Learning [67.58720734177325]
  We propose an unsupervised graph structure learning paradigm, where the learned graph topology is optimized by data itself without any external guidance. Specifically, we generate a learning target from the original data as an "anchor graph", and use a contrastive loss to maximize the agreement between the anchor graph and the learned graph.
  arXiv  Detail & Related papers  (2022-01-17T11:57:29Z)
• Towards Open-World Feature Extrapolation: An Inductive Graph Learning Approach [80.8446673089281]
  We propose a new learning paradigm with graph representation and learning. Our framework contains two modules: 1) a backbone network (e.g., feedforward neural nets) as a lower model takes features as input and outputs predicted labels; 2) a graph neural network as an upper model learns to extrapolate embeddings for new features via message passing over a feature-data graph built from observed data.
  arXiv  Detail & Related papers  (2021-10-09T09:02:45Z)
• Data-Informed Global Sparseness in Attention Mechanisms for Deep Neural Networks [33.07113523598028]
  We propose Attention Pruning (AP), a framework that observes attention patterns in a fixed dataset and generates a global sparseness mask. AP saves 90% of attention computation for language modeling and about 50% for machine translation and GLUE tasks, maintaining result quality.
  arXiv  Detail & Related papers  (2020-11-20T13:58:21Z)

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.