mlsanityData to 7 Oct 2026

ExRole: From Team Trajectories to Executable Roles in Multi-Agent Language Models

arXiv 2608.11949 ↗First posted 12 Aug 2026cs.AIcited by 0

On the map: LLM agents and tool use → General LLM agent frameworks and benchmarks

Abstract (arXiv)

Roles provide an interpretable interface for organizing language-model agents, yet most multi-agent systems treat them as hand-written prompt labels disconnected from learned behavior and parameter updates. We argue that a useful role should instead be an executable control variable: it should summarize behavior predictive of future utility, guide subsequent interaction, and identify the trainable capacity responsible for that behavior. We introduce ExRole, a trajectory-to-role framework that learns future-aware role prototypes from prefix-local team traces, resolves them into readable instructions and token-aligned role markers, and optionally routes shared LoRA rank slots with turn-aligned credit. Across MuSiQue and 2WikiMultiHopQA, ExRole improves over single-agent search by 15.0/14.4 and 13.5/16.1 EM/F1 points, respectively. Against the strongest non-ExRole controls, the corresponding gains remain 11.5/11.6 and 7.7/9.7 points. Across both benchmarks, the controlled results consistently favor trajectory-induced role conditioning over role-free, manual, random, and shuffled alternatives. Role-Agent-Turn interventions further show that the induced roles capture transferable behavioral specialization beyond fixed agent identities or turn positions.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Placed in General LLM agent frameworks and benchmarks, one of 16 directions in LLM agents and tool use. Builds on 13 of its 26 references that are in the map. First posted on arXiv 12 Aug 2026.

Builds on

13 of 26 references.

Closest ancestors · 2

Same direction, newest first

  1. 2025ReSo: A Reward-driven Self-organizing LLM-based Multi-Agent System for Reasoning Tasks4 Mar
  2. 2023CAMEL: Communicative Agents for "Mind" Exploration of Large Language Model Society31 Mar

Other references · 11

  1. 2025Search-R1: Training LLMs to Reason and Leverage Search Engines with Reinforcement Learningcited by 1,371
  2. 2022Interleaving Retrieval with Chain-of-Thought Reasoning for Knowledge-Intensive Multi-Step Questionscited by 911
  3. 2017Counterfactual Multi-Agent Policy Gradientscited by 733
  4. 2023LoraHub: Efficient Cross-Task Generalization via Dynamic LoRA Compositioncited by 359
  5. 2020ROMA: Multi-Agent Reinforcement Learning with Emergent Rolescited by 135
  6. 2020RODE: Learning Roles to Decompose Multi-Agent Taskscited by 125
  7. 2024LoraRetriever: Input-Aware LoRA Retrieval and Composition for Mixed Tasks in the Wildcited by 68
  8. 2025MARFT: Multi-Agent Reinforcement Fine-Tuningcited by 51
  9. 2025Multi-Agent Tool-Integrated Policy Optimizationcited by 16
  10. 2026TRIAGE: Role-Typed Credit Assignment for Agentic Reinforcement Learningcited by 5
  11. 2026Divide and Cooperate: Role-Decomposed Multi-Agent LLM Training with Cross-Agent Learning Signalscited by 2

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2026Role Steering of Language Models for Social Simulations
  2. 2026Role-Agent: Bootstrapping LLM Agents via Dual-Role Evolution
  3. 2026FORTIS: Benchmarking Over-Privilege in Agent Skills
  4. 2026MetaGen: Self-Evolving Roles and Topologies for Multi-Agent LLM Reasoning
  5. 2026Joint Learning of Experiential Rules and Policies for Large Language Model Agents
  6. 2026RoleCDE:Benchmarking and Mitigating Role-Alignment Trade-offs in Role-Playing Agents
  7. 2026Roles with Rails: Contract-Preserving Role Evolution in Multi-Agent Structured Reasoning
  8. 2026Experience Funnel: A State-Policy Alternating Loop for Self-Evolving Agents