GeoPAR: Large-Scale Multi-Agent Combinatorial Optimization with Geometry-Guided Parallel Autoregressive Learning

• URL: http://arxiv.org/abs/2609.00577v1
• Date: Tue, 01 Sep 2026 02:14:05 GMT
• Title: GeoPAR: Large-Scale Multi-Agent Combinatorial Optimization with Geometry-Guided Parallel Autoregressive Learning
• Abstract summary: GeoPAR is a geometry-guided parallel autoregressive reinforcement learning framework for scalable multi-agent optimization.<n>Experiments show that GeoPAR improves large-scale zero-shot generalization while substantially reducing rollout steps and maintaining efficient inference.
• Score: 4.471528509969228
• License: http://creativecommons.org/licenses/by-sa/4.0/
• Abstract: Multi-agent combinatorial optimization problems are notoriously challenging due to their NP-hard nature. Recent parallel autoregressive neural solvers improve inference efficiency by allowing agents to make decisions simultaneously, but their performance often degrades on large-scale instances. This is largely attributable to weak modeling of local geometric structures and the fact that conflicting task selections are handled only after action generation. To address these limitations, we propose GeoPAR, a geometry-guided parallel autoregressive reinforcement learning framework for scalable multi-agent combinatorial optimization. GeoPAR integrates three key components: (1) a projection-window sparse geometry mechanism that builds lightweight local candidate neighborhoods through multi-directional projections, (2) sparse edge-biased attention that injects these geometric relations into node representations, and (3) cache-guided conflict-aware assignment that reuses the geometric cache during decoding to suppress duplicate selections of exclusive tasks. Experiments on heterogeneous vehicle routing and open multi-depot pickup-and-delivery problems show that GeoPAR improves large-scale zero-shot generalization while substantially reducing rollout steps and maintaining efficient inference.

Related papers

• Geometric Self-Supervised Pre-training for Neural Combinatorial Optimization [40.74618635552097]
  We propose a geometric self-supervised pre-training framework for routing graphs.<n>The framework learns robust structural representations prior to the policy optimization phase.<n>The proposed model exhibits remarkable computational efficiency, delivering speedups of up to two orders of magnitude over the exact solver Concorde.
  arXiv  Detail & Related papers  (2026-07-31T20:23:21Z)
• LG-HCC: Local Geometry-Aware Hierarchical Context Compression for 3D Gaussian Splatting [77.81227097905865]
  Anchor-based 3DGS compression schemes reduce gaussian redundancy through some advanced context models.<n>We propose a Local Geometry-aware Hierarchical Context Compression framework for 3DGS that incorporates inter-anchor geometric correlations into anchor pruning and entropy coding for compact representation.<n>Experiments show that LG-HCC effectively alleviates structural preservation issues,achieving superior geometric integrity and rendering fidelity while reducing storage by up to 30.85x compared to the Scaffold-GS baseline on the Mip-NeRF360 dataset.
  arXiv  Detail & Related papers  (2026-03-30T13:39:35Z)
• Decoupling Geometric Planning and Execution in Scalable Multi-Agent Path Finding [44.79409119345322]
  Multi-Agent Path Finding (MAPF) requires collision-free trajectories for multiple agents on a shared graph.<n>We propose a hybrid prioritized framework that separates geometric planning from execution-time conflict resolution.
  arXiv  Detail & Related papers  (2026-03-11T11:04:54Z)
• Tail-Aware Post-Training Quantization for 3D Geometry Models [58.79500829118265]
  Post-Training Quantization (PTQ) enables efficient inference without retraining.<n>PTQ fails to transfer effectively to 3D models due to intricate feature distributions and prohibitive calibration overhead.<n>We propose TAPTQ, a Tail-Aware Post-Training Quantization pipeline for 3D geometric learning.
  arXiv  Detail & Related papers  (2026-02-02T07:21:15Z)
• COMPOSE: Hypergraph Cover Optimization for Multi-view 3D Human Pose Estimation [58.47973015036709]
  3D pose estimation from sparse multi-views is a critical task for action recognition, sports analysis, and human-robot interaction.<n>We propose COMPOSE, a novel framework that formulates multi-view pose correspondence matching as a hypergraph problem.<n> COMPOSE achieves improvements of up to 23% in average precision over previous optimization-based methods and up to 11% over self-supervised end-to-end learned methods.
  arXiv  Detail & Related papers  (2026-01-14T18:50:17Z)
• Multi-Agent Pointer Transformer: Seq-to-Seq Reinforcement Learning for Multi-Vehicle Dynamic Pickup-Delivery Problems [17.3780399150554]
  This paper proposes an end-to-end centralized decision-making framework based on sequence-to-sequence, named Multi-Agent Pointer Transformer (MAPT)<n>MAPT significantly outperforms existing baseline methods in terms of performance and substantial computational time advantages compared to classical operations research methods.
  arXiv  Detail & Related papers  (2025-11-21T17:32:10Z)
• PARCO: Parallel AutoRegressive Models for Multi-Agent Combinatorial Optimization [25.209951743372855]
  PARCO is a reinforcement learning framework designed to construct high-quality solutions for multi-agent tasks efficiently.<n>We evaluate PARCO in multi-agent vehicle routing and scheduling problems, where our approach outperforms state-of-the-art learning methods.
  arXiv  Detail & Related papers  (2024-09-05T17:49:18Z)
• Real-Time Image Segmentation via Hybrid Convolutional-Transformer Architecture Search [51.89707241449435]
  In this paper, we address the challenge of integrating multi-head self-attention into high-resolution representation CNNs efficiently.<n>We develop a multi-target multi-branch supernet method, which fully utilizes the advantages of high-resolution features.<n>We present a series of models via the Hybrid Convolutional-Transformer Architecture Search (HyCTAS) method that searches for the best hybrid combination of light-weight convolution layers and memory-efficient self-attention layers.
  arXiv  Detail & Related papers  (2024-03-15T15:47:54Z)
• Sparse Cross-scale Attention Network for Efficient LiDAR Panoptic Segmentation [12.61753274984776]
  We present SCAN, a novel sparse cross-scale attention network to align multi-scale sparse features with global voxel-encoded attention to capture the long-range relationship of instance context. For the surface-aggregated points, SCAN adopts a novel sparse class-agnostic representation of instance centroids, which can not only maintain the sparsity of aligned features, but also reduce the amount of the network through sparse convolution.
  arXiv  Detail & Related papers  (2022-01-16T05:34:54Z)
• Self-supervised Geometric Perception [96.89966337518854]
  Self-supervised geometric perception is a framework to learn a feature descriptor for correspondence matching without any ground-truth geometric model labels. We show that SGP achieves state-of-the-art performance that is on-par or superior to the supervised oracles trained using ground-truth labels.
  arXiv  Detail & Related papers  (2021-03-04T15:34:43Z)
• DyCo3D: Robust Instance Segmentation of 3D Point Clouds through Dynamic Convolution [136.7261709896713]
  We propose a data-driven approach that generates the appropriate convolution kernels to apply in response to the nature of the instances. The proposed method achieves promising results on both ScanetNetV2 and S3DIS. It also improves inference speed by more than 25% over the current state-of-the-art.
  arXiv  Detail & Related papers  (2020-11-26T14:56:57Z)

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.