Divide, Harmonize, Then Conquer It: Shooting Multi-Commodity Flow Problems with Multimodal Language Models

2026-02-11

AI Summary

This work addresses the trade-off between solution quality and computational efficiency in the multi-commodity flow (MCF) problem by proposing a novel multi-agent framework grounded in multimodal large language models (MLLMs). The approach decomposes the global MCF problem into local subproblems, which are solved in parallel by MLLM-based agents, while multi-agent reinforcement learning ensures global consistency. Innovatively integrating MLLMs into combinatorial optimization, the method introduces a provably convergent in-context gradient descent mechanism that substantially enhances generalization under unseen perturbations—such as link failures or traffic surges. Experimental results on real-world datasets and standard network topologies demonstrate that the proposed framework matches or exceeds the performance of conventional linear programming solvers while achieving 1–2 orders of magnitude faster runtime, with less than 10% performance degradation under dynamic disturbances.

Technology Category

• Multiagent Systems: Multiagent Learning
• Machine Learning: Large Multimodal Models (LMMs)
• Search and Optimization: Mixed Discrete/Continuous Search

Application Category

• Economics, Online Markets and Human Computation: Cost models of using LLMs in production systems
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• User Modeling, Personalization and Recommendation: Large Language Models (LLM) for user modeling and recommendation

Abstract

The multi-commodity flow (MCF) problem is a fundamental topic in network flow and combinatorial optimization, with broad applications in transportation, communication, and logistics, etc. Nowadays, the rapid expansion of allocation systems has posed challenges for existing optimization engines in balancing optimality and tractability. In this paper, we present Pram, the first ML-based method that leverages the reasoning power of multimodal language models (MLMs) for addressing the trade-off dilemma -- a great need of service providers. As part of our proposal, Pram (i) quickly computes high-quality allocations by dividing the original problem into local subproblems, which are then resolved by an MLM-powered"agent", and (ii) ensures global consistency by harmonizing these subproblems via a multi-agent reinforcement learning algorithm. Theoretically, we show that Pram, which learns to perform gradient descent in context, provably converges to the optimum within the family of MCF problems. Empirically, on real-world datasets and public topologies, Pram achieves performance comparable to, and in some cases even surpassing, linear programming solvers (very close to the optimal solution), and substantially lower runtimes (1 to 2 orders of magnitude faster). Moreover, Pram exhibits strong robustness (<10\% performance degradation under link failures or flow bursts), demonstrating MLM's generalization ability to unforeseen events. Pram is objective-agnostic and seamlessly integrates with mainstream allocation systems, providing a practical and scalable solution for future networks.

Problem

• multi-commodity flow
• optimality
• tractability
• network optimization
• allocation systems

Innovation

• multimodal language models
• multi-commodity flow
• multi-agent reinforcement learning
• problem decomposition
• optimization