让语言模型通过密集通信高效协作，实现端到端优化。
Language Model Networks: Supervision-Efficient Learning through Dense Communication
- 用可训练的序列到序列模块替代自然语言通信，实现节点间向量传递。
- 在少量监督下性能接近全监督，训练成本仅增加15%。
- 适合构建多模型协作系统，尤其适合资源受限场景。
语言模型不仅作为独立预测器使用，也越来越多地作为更大推理系统中的组件，如测试时扩展和多智能体协作。本文研究语言模型网络（LMNet），其中预训练语言模型作为可复用节点，智能由拓扑结构、通信方式和优化过程共同产生。现有系统主要通过自然语言通信：部署简单但离散、低效且难以从端任务监督中优化。本文提出一种密集且可微的实现方案——LMNet，使用精简版LLM作为顶点模块，可训练的seq2seq模块作为通信边，使中间节点能交换密集向量，同时在系统边界保持自然语言输入输出。通过跳过中间嵌入与解嵌入步骤，LMNet实现了高效信息传递、端到端梯度优化，并学习超越人工设计协议的通信机制。实验表明，在额外训练成本仅增加约15%的情况下，性能表现优异，且在有限监督下仍具有效适应能力。
原文摘要 · Abstract (English)
Language models are increasingly used not only as standalone predictors but also as components in larger inference systems, from test-time scaling to multi-agent collaboration. We study language model networks, where pre-trained language models serve as reusable nodes and intelligence emerges from their topology, communication, and optimization. Existing systems mostly communicate through natural language: easy to deploy, but discrete, inefficient, and hard to optimize from end-task supervision. We propose LMNet, a dense and differentiable realization of this paradigm. LMNet uses stripped LLMs as vertex modules and trainable seq2seq modules as communication edges, enabling intermediate nodes to exchange dense vectors while preserving natural-language input and output at the system boundary. By bypassing intermediate embedding and de-embedding, LMNet enables efficient information transfer, end-to-end gradient optimization, and learned communication beyond hand-designed protocols. Experiments show performance with small additional training cost and effective adaptation under limited supervision.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。