Router-R1: Teaching LLMs Multi-Round Routing and Aggregation via Reinforcement Learning

• URL: http://arxiv.org/abs/2506.09033v2
• Date: Wed, 18 Jun 2025 16:49:26 GMT
• Title: Router-R1: Teaching LLMs Multi-Round Routing and Aggregation via Reinforcement Learning
• Authors: Haozhen Zhang, Tao Feng, Jiaxuan You,
• Abstract summary: We present textbf generalization-R1, a reinforcement learning framework that formulates multi-LLM routing and aggregation as a sequential decision process.<n>To facilitate learning, we employ a lightweight rule-based reward comprising format rewards, final outcome rewards, and a novel cost reward for optimizing the balance between performance and cost.
• Score: 12.878608250420832
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: The rapid emergence of diverse large language models (LLMs) has spurred the development of LLM routers that assign user queries to the most suitable model. However, existing LLM routers typically perform a single-round, one-to-one mapping (\textit{i.e.}, assigning each query to a single model in isolation), which limits their capability to tackle complex tasks that demand the complementary strengths of multiple LLMs. In this paper, we present \textbf{Router-R1}, a reinforcement learning (RL)-based framework that formulates multi-LLM routing and aggregation as a sequential decision process. Router-R1 instantiates the router itself as a capable LLM, leveraging its reasoning ability to interleave "think" actions (internal deliberation) with "route" actions (dynamic model invocation), and integrates each response into its evolving context. To facilitate learning, we employ a lightweight rule-based reward comprising format rewards, final outcome rewards, and a novel cost reward for optimizing the balance between performance and cost, opening a pathway toward enhancing performance-cost trade-offs via RL. Router-R1 also conditions only on simple model descriptors such as pricing, latency, and example performance, enabling strong generalization to unseen model selection. Experiments on seven general and multi-hop QA benchmarks show that Router-R1 outperforms several strong baselines, achieving superior performance while maintaining robust generalization and cost management.

Related papers

• RCR-Router: Efficient Role-Aware Context Routing for Multi-Agent LLM Systems with Structured Memory [57.449129198822476]
  RCR is a role-aware context routing framework for multi-agent large language model (LLM) systems.<n>It dynamically selects semantically relevant memory subsets for each agent based on its role and task stage.<n>A lightweight scoring policy guides memory selection, and agent outputs are integrated into a shared memory store.
  arXiv  Detail & Related papers  (2025-08-06T21:59:34Z)
• LTRR: Learning To Rank Retrievers for LLMs [53.285436927963865]
  We show that routing-based RAG systems can outperform the best single-retriever-based systems.<n>Performance gains are especially pronounced in models trained with the Answer Correctness (AC) metric.<n>As part of the SIGIR 2025 LiveRAG challenge, our submitted system demonstrated the practical viability of our approach.
  arXiv  Detail & Related papers  (2025-06-16T17:53:18Z)
• RadialRouter: Structured Representation for Efficient and Robust Large Language Models Routing [31.446419903916425]
  Radial is a novel framework for large language models routing.<n>It uses a lightweight Transformer-based backbone with a radial structure named RadialFormer to articulate the query-LLMs relationship.<n>It significantly outperforms existing routing methods by 9.2% and 5.8% in the Balance and Cost First scenarios.
  arXiv  Detail & Related papers  (2025-06-04T12:16:41Z)
• IRT-Router: Effective and Interpretable Multi-LLM Routing via Item Response Theory [26.39979967537193]
  Large language models (LLMs) have demonstrated exceptional performance across a wide range of natural language tasks.<n>While powerful models deliver better results, they come at a high cost, whereas smaller models are more cost-effective but less capable.<n>We propose IRT-Merci, a multi-LLM routing framework that efficiently routes user queries to the most suitable LLM.
  arXiv  Detail & Related papers  (2025-06-01T15:14:58Z)
• Query Routing for Retrieval-Augmented Language Models [38.05904245087491]
  Retrieval-Augmented Generation (RAG) significantly improves the performance of Large Language Models (LLMs) on knowledge-intensive tasks.<n>We observe that external documents dynamically affect LLM's ability to answer queries, while existing routing methods exhibit suboptimal performance in RAG scenarios.<n>We propose RAG, a parametric RAG-aware routing design, which leverages document embeddings and RAG capability embeddings with contrastive learning to capture knowledge representation shifts.
  arXiv  Detail & Related papers  (2025-05-29T03:44:56Z)
• Learning to Route Queries Across Knowledge Bases for Step-wise Retrieval-Augmented Reasoning [60.84901522792042]
  Multimodal Retrieval-Augmented Generation (MRAG) has shown promise in mitigating hallucinations in Multimodal Large Language Models (MLLMs)<n>We propose R1, a novel MRAG framework that learns to decide when and where to retrieve knowledge based on the evolving reasoning state.<n>R1- can adaptively and effectively leverage diverse KBs, reducing unnecessary retrievals and improving both efficiency and accuracy.
  arXiv  Detail & Related papers  (2025-05-28T08:17:57Z)
• Rethinking Predictive Modeling for LLM Routing: When Simple kNN Beats Complex Learned Routers [3.090041654375235]
  We show that a well-tuned k-Nearest Neighbors (kNN) approach outperforms state-of-the-art learned routers across diverse tasks.<n>Our findings reveal that the locality properties of model performance in embedding space enable simple non-parametric methods to achieve strong routing decisions.
  arXiv  Detail & Related papers  (2025-05-19T01:33:41Z)
• How Robust Are Router-LLMs? Analysis of the Fragility of LLM Routing Capabilities [62.474732677086855]
  Large language model (LLM) routing has emerged as a crucial strategy for balancing computational costs with performance.<n>We propose the DSC benchmark: Diverse, Simple, and Categorized, an evaluation framework that categorizes router performance across a broad spectrum of query types.
  arXiv  Detail & Related papers  (2025-03-20T19:52:30Z)
• RouterEval: A Comprehensive Benchmark for Routing LLMs to Explore Model-level Scaling Up in LLMs [45.93874913792025]
  We show a novel model-level scaling up phenomenon in routing large language models (LLMs)<n>This improvement can even surpass the performance of the best single model in the pool and many existing strong LLMs.<n>We introduce RouterEval, a benchmark tailored for router research, which includes over 200,000,000 performance records for 12 popular LLM evaluations.
  arXiv  Detail & Related papers  (2025-03-08T04:07:07Z)
• Universal Model Routing for Efficient LLM Inference [72.65083061619752]
  We consider the problem of dynamic routing, where new, previously unobserved LLMs are available at test time.<n>We propose a new approach to this problem that relies on representing each LLM as a feature vector, derived based on predictions on a set of representative prompts.<n>We prove that these strategies are estimates of a theoretically optimal routing rule, and provide an excess risk bound to quantify their errors.
  arXiv  Detail & Related papers  (2025-02-12T20:30:28Z)
• CITER: Collaborative Inference for Efficient Large Language Model Decoding with Token-Level Routing [56.98081258047281]
  Collaborative Inference with Token-lEvel Routing (CITER) is a framework that enables efficient collaboration between small and large language models.<n>We formulate router training as a policy optimization, where the router receives rewards based on both the quality of predictions and the inference costs of generation.<n>Our experiments show that CITER reduces the inference costs while preserving high-quality generation, offering a promising solution for real-time and resource-constrained applications.
  arXiv  Detail & Related papers  (2025-02-04T03:36:44Z)
• Glider: Global and Local Instruction-Driven Expert Router [83.785832410832]
  "Model MoErging" methods prioritize generalization to unseen tasks at the expense of performance on held-in tasks. We propose Global and Local Instruction Driven Expert Router (GLIDER) that integrates a multi-scale routing mechanism. GLIDER achieves substantially improved held-in performance while maintaining strong generalization on held-out tasks.
  arXiv  Detail & Related papers  (2024-10-09T17:59:14Z)
• ArCHer: Training Language Model Agents via Hierarchical Multi-Turn RL [80.10358123795946]
  We develop a framework for building multi-turn RL algorithms for fine-tuning large language models. Our framework adopts a hierarchical RL approach and runs two RL algorithms in parallel. Empirically, we find that ArCHer significantly improves efficiency and performance on agent tasks.
  arXiv  Detail & Related papers  (2024-02-29T18:45:56Z)

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.