SLMs as Multi-Agent Routers: A Progressive SFT and Reinforcement Learning Approach

• URL: http://arxiv.org/abs/2608.00030v1
• Date: Wed, 15 Jul 2026 11:11:39 GMT
• Title: SLMs as Multi-Agent Routers: A Progressive SFT and Reinforcement Learning Approach
• Abstract summary: Specialised retrieval agents typically surface higher quality results than general-purpose search.<n>Current approaches route queries based on inferred topic or intent, but it does not incorporate signal from retrieved content.<n>We address this by training a small language model via supervised fine-tuning followed by reinforcement learning.
• Score: 35.31208793946578
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Specialised retrieval agents typically surface higher quality results than general-purpose search, but selecting the optimal agent for a given query remains an open problem. Current approaches route queries based on inferred topic or intent, however intent-based selection is fundamentally limited: it does not incorporate signal from retrieved content, and cannot detect when a topically aligned agent produces low-relevance results. We address this by training a small language model via supervised fine-tuning followed by reinforcement learning to jointly perform agent selection and structured parameter generation for downstream tool calls, using a hierarchical reward function grounded in retrieval relevance along with query-agent topic alignment. This enables the model to learn task-dependent agent suitability from retrieval performance: which agents reliably yield high-relevance results for which query distributions, and when to redirect queries away from specialised agents despite surface-level topical overlap. On a targeted subset of such agent-query mismatches, the trained model achieves an NDCG@10 of 0.918 compared to 0.539 and 0.490 for two LLM baselines (Amazon Nova Lite and Claude Haiku 4.5) that route on intent alone. Overall, it achieves a mean NDCG@10 of 0.771 (+0.177 over Nova Lite, +0.219 over Haiku) with a mean selection latency of 120.1ms, an 82.4% reduction over Nova Lite.

Related papers

• From Sequence to Structure: Relational Uncertainty Propagation for LLM Agents [75.69180947909148]
  We propose a trajectory-level UQ framework for large language model (LLM) agents.<n>RuPA represents an execution history as a directed trajectory graph in which reasoning states, tool interactions, and environment feedback are nodes connected by temporal and semantic dependency edges.<n>We evaluate RUPA on representative agent benchmarks, including $$-2, Terminal-Bench-2, and GAIA, using 6 open-source LLMs spanning multiple model families.
  arXiv  Detail & Related papers  (2026-08-17T01:40:14Z)
• RouteRec: Strict Evaluation of Recommender-Agent Selection and Aggregation [4.548136909830987]
  RouteRec is a framework that compares request-level hard selection with item-level learned aggregation.<n>On MovieLens-1M, the full quality oracle has substantial headroom (HR@10 = 0.584)<n>The resulting lesson is not that routing is solved, but item-level aggregation is the more promising action space.
  arXiv  Detail & Related papers  (2026-07-10T18:58:52Z)
• When Should LLMs Search? Counterfactual Supervision for Search Routing [0.04144015562528761]
  We consider whether search is needed to improve task success relative to a no-search execution.<n>We construct an oracle over NO SEARCH, SEARCH, and UNSOLVED based on task-specific success.<n>We train search-routing policies with supervised fine-tuning and preference optimization.
  arXiv  Detail & Related papers  (2026-07-07T02:23:12Z)
• AgentIR: Reasoning-Aware Retrieval for Deep Research Agents [76.29382561831105]
  Deep Research agents generate explicit natural language reasoning before each search call.<n> Reasoning-Aware Retrieval embeds the agent's reasoning trace alongside its query.<n>DR- Synth generates Deep Research retriever training data from standard QA datasets.<n>AgentIR-4B achieves 68% accuracy with the open-weight agent Tongyi-DeepResearch.
  arXiv  Detail & Related papers  (2026-03-04T18:47:26Z)
• AgentSelect: Benchmark for Narrative Query-to-Agent Recommendation [39.61543921719145]
  AgentSelect is a benchmark that reframes agent selection as narrative query-to-agent recommendation.<n>It converts heterogeneous evaluation artifacts into unified, positive-only interaction data.<n>AgentSelect provides the first unified data and evaluation infrastructure for agent recommendation.
  arXiv  Detail & Related papers  (2026-03-04T06:17:51Z)
• Learning to Refine: An Agentic RL Approach for Iterative SPARQL Query Construction [0.18907108368038208]
  Current methods lack the adaptive policies needed to dynamically debug queries based on real-time execution feedback.<n>This paper introduces a novel agentic framework where an LLM learns a resilient policy for the sequential process of iterative SPARQL construction.<n>We show that a compact 3B- parameter model, trained exclusively via outcome-driven Reinforcement Learning (GRPO), can learn effective policies for this task.
  arXiv  Detail & Related papers  (2025-11-14T08:44:58Z)
• Compliance Brain Assistant: Conversational Agentic AI for Assisting Compliance Tasks in Enterprise Environments [2.8724171056550256]
  Compliance Brain Assistant (CBA) is a conversational, agentic AI assistant designed to boost the efficiency of daily compliance tasks for personnel in enterprise environments.<n>To strike a good balance between response quality and latency, we design a user query router that can intelligently choose between FastTrack mode and FullAgentic mode.
  arXiv  Detail & Related papers  (2025-07-23T07:51:10Z)
• MALT: Improving Reasoning with Multi-Agent LLM Training [67.76186488361685]
  MALT (Multi-Agent LLM Training) is a novel post-training strategy that divides the reasoning process into generation, verification, and refinement steps.<n>On MATH, GSM8K, and CSQA, MALT surpasses the same baseline LLM with a relative improvement of 15.66%, 7.42%, and 9.40% respectively.
  arXiv  Detail & Related papers  (2024-12-02T19:30:36Z)
• Large Language Models are Effective Text Rankers with Pairwise Ranking Prompting [65.00288634420812]
  Pairwise Ranking Prompting (PRP) is a technique to significantly reduce the burden on Large Language Models (LLMs) Our results are the first in the literature to achieve state-of-the-art ranking performance on standard benchmarks using moderate-sized open-sourced LLMs.
  arXiv  Detail & Related papers  (2023-06-30T11:32:25Z)
• Reinforcement Learning for Branch-and-Bound Optimisation using Retrospective Trajectories [72.15369769265398]
  Machine learning has emerged as a promising paradigm for branching. We propose retro branching; a simple yet effective approach to RL for branching. We outperform the current state-of-the-art RL branching algorithm by 3-5x and come within 20% of the best IL method's performance on MILPs with 500 constraints and 1000 variables.
  arXiv  Detail & Related papers  (2022-05-28T06:08:07Z)

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.