AI Paper Atlas

Research map / Multi-agent systems and games / LLM agents

GraphPlanner: Graph-Based Agentic Routing for LLMs

Tao Feng, Haozhen Zhang, Zijie Lei, Peixuan Han, Jiaxuan You

ICLR 2026 · Multi-agent systems and games › LLM agents · approach: Robust and efficient agent execution

Read the paperCheck your idea against 137,217 papers

AI-extracted summary. Written by AI from the paper's full text. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Proposes GraphPlanner, a graph-based agentic router that decides which LLM or agent to call at each step of multi-agent workflows, extending LLM routing to agentic settings.

Problem

LLM routing has mostly targeted single-turn queries; agentic settings requiring task planning, multi-round cooperation among heterogeneous agents, and memory use lack suitable routing methods.

Approach

Models routing for agentic workflows as a heterogeneous graph over queries, agents and workflow state, so a learned policy can select agents at each step while balancing performance and cost. Details beyond the abstract are not available.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
use agentic workflow router GraphPlanner instead of best baseline routerLLM routing · Phase 2 average accuracyaccuracy: 54.3% -> 63.6%Improved
use GraphPlanner instead of Router-R1 on costLLM routing · Phase 2 average costcost: 76.3 -> 605.0Hurt
use GraphPlanner to pick backbones inside fixed workflowsPhase 1 workflows · Phase 1 Depth=1 Width=3 averageaccuracy: 54.80% -> 58.60%Improved
generate query-specific workflows (Phase 2) instead of optimizing within fixed workflows (Phase 1)GraphPlanner · average accuracyaccuracy: 58.6 -> 63.6Improved
use GraphPlanner on commonsense reasoning (Phase 2)LLM routing · Phase 2 CSaccuracy: 82.7% -> 78.0%No clear effect
add historical memory to the router (ablation)GraphPlanner · Phase 2 five scenariosaccuracyImproved
use heterogeneous GARNet instead of homogeneous GNN (ablation)GraphPlanner · Phase 2 five scenariosaccuracyImproved
use transductive inference (reuse stored training interactions) instead of inductive (ablation)GraphPlanner · Phase 2 five scenariosaccuracyImproved
add Thinker and Verifier roles with training (ablation)GraphPlanner · Phase 2 Mathaccuracy: 67.0% -> 70.5%Improved
offer unseen roles at test time with no training (ablation)GraphPlanner · Phase 2 Mathaccuracy: 67.0% -> 68.5%Improved
apply GraphPlanner to unseen tasks zero-shotGraphPlanner · LogicGrid, MGSM, CommonGenaccuracy: 58% -> 78%Improved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy