Inherit-MAS: Test-Time Evolution of Multi-Agent Systems through Workflow and Execution Inheritance

Hugging Face Daily Papers Papers
multi-agent-systems test-time-evolution workflow-inheritance large-language-models execution-reuse agent-optimization

Summary

Inherit-MAS 通过在工作流和执行两个层面引入显式继承机制，实现多智能体系统在测试时的演化优化，在 WorkBench 和 HotpotQA 上超越了 EvoAgent、EvoMAS、TacoMAS 等基线，并显著降低了 token 消耗。

Multi-agent systems (MAS) built from large language models coordinate specialized agents to tackle complex tasks, but effective workflows are difficult to design in advance. Test-time evolution refines workflows using execution feedback, yet broad revisions can disturb useful components, while re-executing unchanged requests can incur redundant computation. Inspired by the interplay of inheritance and selection in biological evolution, we introduce Inherit-MAS, which makes inheritance explicit at the workflow and execution levels. A meta-model first synthesizes a workflow of worker agents with declared roles, communication inputs, and tool permissions, and a separately prompted judge scores each executed candidate and diagnoses its deficiencies. In ordinary refinement rounds, workflow inheritance starts from the latest completed candidate, may discard removable nodes judged unhelpful, and applies a validated edit to address the diagnosed deficiency. When the new candidate executes, execution inheritance inherits eligible stored results only if the complete resolved request and execution context match, avoiding redundant model and tool calls. With GPT-4o-mini workers, Inherit-MAS achieves 55.4\% completion on WorkBench and 49.7\% joint F1 on HotpotQA FullWiki, outperforming EvoAgent, EvoMAS, and TacoMAS. With Qwen3-32B workers, it also exceeds these evolving-MAS baselines on both benchmarks. Compared with rerunning the same controller with execution inheritance disabled, execution inheritance reduces worker-token usage by 29.1\% on WorkBench and 34.6\% on HotpotQA, and total token usage by 5.3\% and 18.1\%.
Original Article
View Cached Full Text

Cached at: 10/09/26, 06:30 AM

Paper page - Inherit-MAS: Test-Time Evolution of Multi-Agent Systems through Workflow and Execution Inheritance

Source: https://huggingface.co/papers/2610.02396

Abstract

Multi-agentsystems(MAS)builtfromlargelanguagemodelscoordinatespecializedagentstotacklecomplextasks,buteffectiveworkflowsaredifficulttodesigninadvance.Test-timeevolutionrefinesworkflowsusingexecutionfeedback,yetbroadrevisionscandisturbusefulcomponents,whilere-executingunchangedrequestscanincurredundantcomputation.Inspiredbytheinterplayofinheritanceandselectioninbiologicalevolution,weintroduceInherit-MAS,whichmakesinheritanceexplicitattheworkflowandexecutionlevels.Ameta-modelfirstsynthesizesaworkflowofworkeragentswithdeclaredroles,communicationinputs,andtoolpermissions,andaseparatelypromptedjudgescoreseachexecutedcandidateanddiagnosesitsdeficiencies.Inordinaryrefinementrounds,workflowinheritancestartsfromthelatestcompletedcandidate,maydiscardremovablenodesjudgedunhelpful,andappliesavalidatededittoaddressthediagnoseddeficiency.Whenthenewcandidateexecutes,executioninheritanceinheritseligiblestoredresultsonlyifthecompleteresolvedrequestandexecutioncontextmatch,avoidingredundantmodelandtoolcalls.WithGPT-4o-miniworkers,Inherit-MASachieves55.4\%completiononWorkBenchand49.7\%jointF1onHotpotQAFullWiki,outperformingEvoAgent,EvoMAS,andTacoMAS.WithQwen3-32Bworkers,italsoexceedstheseevolving-MASbaselinesonbothbenchmarks.Comparedwithrerunningthesamecontrollerwithexecutioninheritancedisabled,executioninheritancereducesworker-tokenusageby29.1\%onWorkBenchand34.6\%onHotpotQA,andtotaltokenusageby5.3\%and18.1\%.

View arXiv pageView PDFGitHub0Add to collection

Get this paper in your agent:

hf papers read 2610\.02396

Don’t have the latest CLI?curl \-LsSf https://hf\.co/cli/install\.sh \| bash

Models citing this paper0

No model linking this paper

Cite arxiv.org/abs/2610.02396 in a model README.md to link it from this page.

Datasets citing this paper0

No dataset linking this paper

Cite arxiv.org/abs/2610.02396 in a dataset README.md to link it from this page.

Spaces citing this paper0

No Space linking this paper

Cite arxiv.org/abs/2610.02396 in a Space README.md to link it from this page.

Collections including this paper0

No Collection including this paper

Add this paper to acollectionto link it from this page.

Similar Articles

Inherit-MAS: Test-Time Evolution of Multi-Agent Systems through Workflow and Execution Inheritance

arXiv cs.LG

The paper introduces Inherit-MAS, a test-time evolution framework for LLM multi-agent systems that makes workflow and execution inheritance explicit, using a meta-model to synthesize workflows and a judge to refine them. With GPT-4o-mini and Qwen3-32B workers, it outperforms EvoAgent, EvoMAS, and TacoMAS on WorkBench and HotpotQA while cutting worker-token usage by up to 34.6%.

EvoMAS: Learning Execution-Time Workflows for Multi-Agent Systems

arXiv cs.AI

EvoMAS is a framework for learning execution-time workflows in multi-agent systems by formulating workflow construction as a sequential decision problem. It outperforms static multi-agent design methods on complex tasks by adapting agent coordination dynamically based on evolving task states.

TMAS: Scaling Test-Time Compute via Multi-Agent Synergy

Hugging Face Daily Papers

TMAS introduces a multi-agent framework that enhances large language model reasoning by scaling test-time compute through structured collaboration and hierarchical memory systems. The approach uses specialized agents, cross-trajectory information flow, and hybrid reward reinforcement learning to improve iterative scaling and stability on challenging reasoning benchmarks.

Recursive Multi-Agent Systems

Papers with Code Trending

This paper introduces RecursiveMAS, a framework that extends recursive scaling principles to multi-agent systems for improved collaborative reasoning efficiency and accuracy. It demonstrates significant speedups and token reduction across various benchmarks compared to standard baselines.