Localized LoRA-MoE: Block-wise Low-Rank Experts With Adaptive Routing

• URL: http://arxiv.org/abs/2607.05114v1
• Date: Mon, 06 Jul 2026 14:06:18 GMT
• Title: Localized LoRA-MoE: Block-wise Low-Rank Experts With Adaptive Routing
• Abstract summary: We introduce Localized LoRA-MoE, a unified framework that fuses localized spatial blocking with dynamic, context-efficient routing.<n>Our empirical results establish that decentralized cell-level gating achieves complete statistical parity with an omniscient global coordinator.
• Score: 12.174849960744211
• License: http://creativecommons.org/licenses/by-nc-nd/4.0/
• Abstract: Large Language Models (LLMs) and high-dimensional perception networks increasingly rely on parameter-efficient fine-tuning (PEFT) to adapt to diverse operational contexts. However, standard methods like LoRA are structurally limited by a monolithic bottleneck, making them highly susceptible to gradient warfare. Interleaved multi-task streams may trigger destructive optimization feedback, collapsing adapter weights into unspecialized averages. While recent spatial partitioning methods have introduced block-wise isolation, they remain trapped in static topologies, unable to adapt to dynamic task-switching or environmental sensor failure. In this work, we introduce Localized LoRA-MoE, a unified framework that fuses localized spatial blocking with dynamic, context-conditioned routing. We propose and evaluate two novel architectural paradigms: Block-Wise LoRA-MoE (Centralized Macro-Routing), which modulates the entire structural grid via a monolithic context signal, and Cell-Wise LoRA-MoE (Decentralized Micro-Routing), which empowers every coordinate cell in the matrix grid with autonomous, localized expert gating. Through a comprehensive suite of benchmarks, ranging from high-dimensional SVD matrix simulations and real-world tabular transformations to spatial vision perception under sensor degradation, we demonstrate that both architectures resolve optimization deadlocks inherent in static baselines. Our empirical results establish that decentralized cell-level gating achieves complete statistical parity with an omniscient global coordinator, providing a robust "gradient firewall" that protects surviving pathways from fault-propagated corruption. Our proposals consistently outperform static baselines, offering a scalable and parameter-efficient solution for dynamic model adaptation across granular coordinate fields and shifting operational regimes.

Related papers

• Hierarchical Federated Learning with Dynamic Clustering and Adaptive Regularization for Robust Infrastructure Inspection [46.657161089480375]
  This paper proposes a novel hierarchical federated learning framework.<n>The framework orchestrates a synergistic two-tier optimization strategy.<n> Comprehensive evaluations on a large-scale, real-world structural inspection dataset demonstrate that the hierarchical integration of macro-clustering and micro-regularization successfully neutralizes dual-level heterogeneity.
  arXiv  Detail & Related papers  (2026-06-02T03:14:26Z)
• CoRAL: Contact-Rich Adaptive LLM-based Control for Robotic Manipulation [4.600772894287281]
  Large Language Models (LLMs) and Vision-Language Models (VLMs) demonstrate remarkable capabilities in high-level reasoning and semantic understanding.<n>We propose CoRAL, a modular framework that enables zero-shot planning by decoupling high-level reasoning from low-level control.
  arXiv  Detail & Related papers  (2026-05-04T13:49:19Z)
• ATRS: Adaptive Trajectory Re-splitting via a Shared Neural Policy for Parallel Optimization [14.810654249835572]
  We propose ATRS, a novel framework that embeds a shared Deep Reinforcement Learning policy into the parallel ADMM loop.<n>We show that ATRS convergence accelerates by up to 26.2% and the computation time by up to 19.1%.<n>Real-world experiments further confirm its applicability to both large-scale offline global planning and real-time onboard replanning within 35 ms per cycle.
  arXiv  Detail & Related papers  (2026-04-24T16:58:14Z)
• AnchorRefine: Synergy-Manipulation Based on Trajectory Anchor and Residual Refinement for Vision-Language-Action Models [60.04879435087352]
  Vision-language-action (VLA) policies generate actions within a single unified space.<n>We propose AnchorRefine, a hierarchical framework that factorizes VLA action modeling into trajectory anchor and residual refinement.<n> Experiments on LIBERO, CALVIN, and real-robot tasks demonstrate that AnchorRefine consistently improves both regression-based and diffusion-based VLA backbones.
  arXiv  Detail & Related papers  (2026-04-20T04:25:24Z)
• Bypassing the CSI Bottleneck: MARL-Driven Spatial Control for Reflector Arrays [8.046649921554472]
  This paper presents a fully autonomous Multi-Agent Reinforcement Learning (MARL) framework to control mechanically adjustable metallic reflector arrays.<n>High-fidelity ray-tracing simulations in dynamic non-line-of-sight (NLOS) environments demonstrate that this multi-agent approach rapidly adapts to user mobility.<n>These results validate the efficacy of MARL-driven spatial abstractions as a scalable, highly practical pathway toward AI-empowered wireless networks.
  arXiv  Detail & Related papers  (2026-04-06T20:47:50Z)
• Enhancing Gradient Inversion Attacks in Federated Learning via Hierarchical Feature Optimization [56.95448807869383]
  Federated Learning (FL) has emerged as a compelling paradigm for privacy-preserving distributed machine learning.<n>Recent studies find that the gradients exchanged in the FL system are also vulnerable to privacy leakage.<n>We propose textbfGradient textbfInversion over textbfFeature textbfDomains (GIFD)
  arXiv  Detail & Related papers  (2026-04-01T14:32:15Z)
• OmniVL-Guard: Towards Unified Vision-Language Forgery Detection and Grounding via Balanced RL [63.388513841293616]
  Existing forgery detection methods fail to handle the interleaved text, images, and videos prevalent in real-world misinformation.<n>To bridge this gap, this paper targets to develop a unified framework for omnibus vision-language forgery detection and grounding.<n>We propose textbf OmniVL-Guard, a balanced reinforcement learning framework for omnibus vision-language forgery detection and grounding.
  arXiv  Detail & Related papers  (2026-02-11T09:41:36Z)
• MAESTRO: Meta-learning Adaptive Estimation of Scalarization Trade-offs for Reward Optimization [56.074760766965085]
  Group-Relative Policy Optimization has emerged as an efficient paradigm for aligning Large Language Models (LLMs)<n>We propose MAESTRO, which treats reward scalarization as a dynamic latent policy, leveraging the model's terminal hidden states as a semantic bottleneck.<n>We formulate this as a contextual bandit problem within a bi-level optimization framework, where a lightweight Conductor network co-evolves with the policy by utilizing group-relative advantages as a meta-reward signal.
  arXiv  Detail & Related papers  (2026-01-12T05:02:48Z)
• TC-LoRA: Temporally Modulated Conditional LoRA for Adaptive Diffusion Control [15.019368747393862]
  Current controllable diffusion models rely on fixed architectures that modify intermediate activations to inject guidance conditioned on a new modality.<n>We introduce TC-LoRA, a new paradigm that enables dynamic, context-aware control by conditioning the model's weights directly.
  arXiv  Detail & Related papers  (2025-10-10T17:13:02Z)
• Towards Efficient General Feature Prediction in Masked Skeleton Modeling [59.46799426434277]
  We propose a novel General Feature Prediction framework (GFP) for efficient mask skeleton modeling.<n>Our key innovation is replacing conventional low-level reconstruction with high-level feature prediction that spans from local motion patterns to global semantic representations.
  arXiv  Detail & Related papers  (2025-09-03T18:05:02Z)

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.