A Responsibility-Centered Coordination Mechanism for Decentralized Multi-Agent Systems

Thi-Nhung Nguyen; Viet-Thanh Pham; Linhao Luo; Thuy-Trang Vu; Dinh Phung

cs.MAOct 15, 2026

A Responsibility-Centered Coordination Mechanism for Decentralized Multi-Agent Systems

Authors: Thi-Nhung Nguyen, Viet-Thanh Pham, Linhao Luo, Thuy-Trang Vu, Dinh Phung

Organizations: Department of Data Science & AI, Monash University

Abstract

Centralized orchestration provides comprehensive multi-agent coordination but creates bottlenecks in scalability, adaptability, and collective performance. Decentralized approaches alleviate these bottlenecks by distributing coordination across agents, yet often leave responsibility for delegated dependencies and failure recovery unclear. To address this gap, we propose RCM, a responsibility-centered coordination mechanism that leverages commitments to couple task delegation with explicit responsibilities of the involved parties. Through recursive delegation, agents retain their upstream responsibilities while locally orchestrating downstream commitments, thereby preserving accountability across delegation chains, creating a network of local orchestration. We operationalize this mechanism locally through responsibility-constrained decision-making and a responsibility-oriented communication protocol (RCP), enabling agents to autonomously establish, manage, and terminate commitments. As agents act in parallel, their local actions continuously evolve the commitment graph, allowing global coordination to emerge and adapt toward a shared mission without centralized control. Experiments across reasoning, coding, and executable agent environments demonstrate that RCM outperforms strong centralized and decentralized coordination baselines by up to 18.7 points in task performance and 10.2 points in Responsibility, while improving failure and harm detection without task-specific training.

Figures & tables

Figure 1: Overview of RCM . The right side shows global coordination through a dynamic network of commitments and local failure recovery. The left side illustrates each agent’s internal observe–decide–act loop, through which local decisions collectively give rise to this global coordination.

Figure 2: Example commitment handshake between a delegator and a delegatee. Blue arrows show commitment state updates; red dashed arrows show the delegatee’s acceptance, progress reports, and outcome with completion evidence.

RCP operation	Initiator and semantics
delegate.propose	$a_{i}\to a_{j}$ : open a delegation; record the $g,\tau$ proposal as Requested .
delegate.accept / decline	$a_{j}\to a_{i}$ : accept or decline the proposal version; acceptance establishes commitment edge $C$ in Agreed .
commitment.query / communicate	$a_{i}\leftrightarrow a_{j}$ : exchange questions, clarifications, information, or intervention requests within scope; correlate responses by reply_to .
commitment.notify	$a_{i}\leftrightarrow a_{j}$ : report consequential changes; $a_{j}$ also reports start, progress, partial results, and exceptions.
outcome.submit / verify	$a_{j}\to a_{i}$ : submit outcome evidence; $a_{i}$ verifies it against $\beta$ before discharge.
commitment.revise	$a_{i}\leftrightarrow a_{j}$ : counter or revise $g,\tau$ ; changes take effect only after agreement on the new version.

Table 1: Core RCP operations between delegator

a_{i}

and delegatee

a_{j}

. Messages inform the recipient’s decisions under its active responsibility in

\rho_{C}

; lifecycle updates are guarded changes to commitment edges.

Workflow

Method

Dynamic workflow

Responsi- bility

Distributed M/R

Qwen3-4B

GPT-5.4

Non-Agentic

Agentic

Non-Agentic

Agentic

AIME25

HumanEval

Claw-Eval

Claw-Live

AIME25

HumanEval

Claw-Eval

Claw-Live

Broadcast

MAS-Broadcast

\times

Unspecified

\times

30.0

\pm

3.3

83.3

\pm

2.2

35.7

\pm

1.9

40.2

\pm

3.7

43.3

\pm

8.4

93.9

\pm

1.5

50.5

\pm

1.4

64.4

\pm

2.7

Designed

MachineSoM

\times

Debate

\times

16.7

\pm

5.1

87.9

\pm

0.9

41.6

\pm

1.4

56.3

\pm

1.5

66.7

\pm

0.0

90.9

\pm

4.0

44.2

\pm

1.5

51.5

\pm

3.1

Sparse MAD

\times

Debate

\times

6.7

\pm

3.3

81.8

\pm

2.3

43.0

\pm

0.5

59.5

\pm

0.8

60.0

\pm

10.2

98.5

\pm

0.0

46.9

\pm

0.9

60.0

\pm

1.1

MetaGPT

\times

Role

\times

36.7

\pm

8.8

86.4

\pm

1.5

40.5

\pm

1.9

52.4

\pm

1.0

63.3

\pm

5.1

89.4

\pm

3.5

51.4

\pm

2.6

57.6

\pm

2.4

Table 2: Performance across tasks, environments, and executor models. Dynamic workflow indicates whether task relationships can change during execution. Responsibility is unspecified under broadcast, induced through debate or role assignment, implicit in workflow structure, execution-based, or explicit through commitment. Distributed M/R denotes distributed monitoring and recovery.

\checkmark

\circ

, and

\times

indicate full, partial, and no support.

Figure 3: Agent scaling on Claw-Eval .

Broadcast
Method	Comm. $\uparrow$	Mon. $\uparrow$	Intv. $\uparrow$	Ful. $\uparrow$	Resp. $\uparrow$
MAS-Broadcast	82.9 $\pm$ 5.0	89.7 $\pm$ 4.7	44.8 $\pm$ 5.2	38.1 $\pm$ 1.7	63.9 $\pm$ 2.7
Designed
MachineSoM	94.7 $\pm$ 6.0	77.8 $\pm$ 4.8	13.9 $\pm$ 5.1	30.2 $\pm$ 1.5	54.1 $\pm$ 1.5
Sparse MAD	93.9 $\pm$ 0.4	77.4 $\pm$ 1.3	12.9 $\pm$ 1.4	33.7 $\pm$ 1.1	54.5 $\pm$ 0.9
MetaGPT	75.4 $\pm$ 10.0	83.7 $\pm$ 7.7	51.2 $\pm$ 10.1	39.8 $\pm$ 2.9	62.5 $\pm$ 5.2

Table 3: Responsibility metrics on Claw-Eval with GPT-5.4. Resp. is the equal-weight mean of Comm., Mon., Intv., and Ful.

Trained baselines (Trained $\checkmark$ )
Model	Pair	Agent	Error	Avg.
DCL (Aegis)	1.60	8.40	14.67	8.22
Aegis-SFT (Qwen2.5-14B)	4.03	51.14	9.87	21.68
Aegis-GRPO (Qwen2.5-14B)	2.45	54.43	4.15	20.34
Qwen3-8B-Thinking + GRPO	8.10	53.12	11.25	24.16
Training-free baselines (Trained $\times$ )

Table 4: Failure attribution on Who&When under the Aegis protocol. Baseline results are from Kong et al. (2026) . Avg. is the mean of Pair, Agent, and Error micro-F1.

Guard models (Trained $\checkmark$ )
Model	Acc.	Prec.	Rec.	F1
LlamaGuard3-8B	53.1	85.7	3.8	7.3
LlamaGuard4-12B	58.1	63.8	30.9	41.7
Qwen3-Guard	51.5	40.0	0.4	0.8
ShieldAgent	62.5	58.0	81.4	67.7
AgentDoG-Qwen3-4B	64.0	59.2	88.9	71.1

Table 5: Harmful-trajectory detection on ATBench-C. Baseline results are from Li et al. (2026b) .

Variant	Task $\uparrow$	Resp. $\uparrow$
Full RCM	72.8 $\pm$ 1.3	82.8 $\pm$ 0.0
Relevant-1 initial receiver	61.5 $\pm$ 1.6	73.9 $\pm$ 5.0
Random-1 initial receiver	59.1 $\pm$ 1.4	67.5 $\pm$ 3.6
Without explicit delegator responsibility	60.4 $\pm$ 1.7	68.5 $\pm$ 1.6
Without explicit delegatee responsibility	50.4 $\pm$ 1.8	65.0 $\pm$ 0.5
Without commitment lifecycle	56.6 $\pm$ 1.4	47.8 $\pm$ 0.4

Table 6: Ablation on Claw-Eval with GPT-5.4. Relevant-1 and Random-1 initially assign the task to the most relevant or a randomly selected agent, respectively, instead of all agents.

Emergent task architectures
Observed behavior	Rate (%)
Linear / sequential support	35.2
Hybrid network	27.5
Specialist hub	24.2
Delegation star	9.9
Hierarchical cascade	3.3

Table 7: Emergent self-organization and local coordination. Complexity bounds are structural, with

d

events per edge.

Appendix figures & tables3 assets

Supplementary material from the paper’s appendix.

Appendix

Status	Description
REQUESTED	The delegator records the proposed goal and terms and sends the request. No obligation exists yet.
AGREED	The delegatee accepts the proposal. The commitment is now binding, although its activation conditions may not yet hold.
DETACHED	The activation conditions in $\tau$ hold, making the obligation to achieve $g$ active.
EXECUTING	Records that the delegatee has started work and reported progress.
DISCHARGED	The delegator has verified that the outcome satisfies the agreed completion criteria.
VIOLATED	The delegator has verified that a due outcome was not achieved or that an agreed constraint was breached.

Appendix

Table 8: Commitment lifecycle states and their operational semantics in RCM .

Figure 4: Commitment lifecycle and state transitions relevant to responsibility in RCM .

Dataset	Setting	Primary metric
AIME25	Mathematical reasoning	Accuracy
HumanEval	Code generation	Pass@1
Claw-Eval	Executable agent tasks	Score, completion, safety, responsibility
Claw-Eval-Live	Dynamic executable tasks	Score

Appendix

Table 9: Evaluation benchmarks, task settings, and primary metrics.

Multi-Agent Coordination

Multi-agent coordination studies how autonomous agents align actions, share information, and divide work to achieve individual or collective goals.

See all 517 papers

Distilling Collaborative Dynamics into Latent Space for Implicit Coordination in Decentralized Multi-Agent Manipulation
Jun 22, 2026Chanyoung Park, Minsung Yoon, Andrew Jeong +1Multi-Agent Coordination Centralized Training with Decentralized Execution
School of Computing at the Korea Advanced Institute of Science and Technology (KAIST), Daejeon, 34141, Republic of Korea
Multi-Agent Coordination Adaptation via Structure-Guided Orchestration
May 25, 2026Haoran Li, Shulun Chen, Shaoyuan Sun +1Multi-Agent LLM Systems Multi-Agent System Optimization
Nanjing University · University of Technology Sydney, Sydney, Australia · University of New South Wales, Sydney, Australia
Optimization without Future Compromises? Decentralized Coordination via Collective and Reinforcement Learning
Sep 22, 2025Chuhao Qin, Evangelos PournarasMulti-Agent System Optimization Multi-Agent Coordination
School of Computer Science, University of Leeds, UK

Jun 22, 2026J/K move · Enter open · S save

Distilling Collaborative Dynamics into Latent Space for Implicit Coordination in Decentralized Multi-Agent Manipulation

Chanyoung Park, Minsung Yoon, Andrew Jeong, Sung-eui Yoon

School of Computing at the Korea Advanced Institute of Science and Technology (KAIST), Daejeon, 34141, Republic of Korea

Multi-Agent Coordination Centralized Training with Decentralized Execution Multi-Agent Systems

Open paper AskPDF arXiv

Multi-arm manipulation demands precise spatiotemporal coordination, yet many centralized approaches scale poorly as team size increases. To address this, we propose CLS-DP, a decentralized multi-agent framework that enables implicit coordination under partial observability without shared global views, explicit state information, or inter-agent communication. Under the centralized training and decentralized execution (CTDE) paradigm, CLS-DP distills privileged multi-agent dynamics into a latent space. At deployment, each agent infers a collaborative latent from its local RGB observation and a shared task instruction; it then conditions the diffusion denoising process on this latent. This design enables implicit coordination with a per-agent cost independent of team size. Across six RoboFactory benchmark tasks spanning two to four agents, CLS-DP achieves a 38% mean success rate, outperforming the best centralized baseline (20%) and a decentralized ablation without the collaborative latent (9%). It also maintains superior parameter efficiency across all agent configurations. Attribution maps show that an agent conditioned on the collaborative latent places high attribution on the joints and grippers of both itself and its teammates throughout execution. This suggests that the learned latent efficiently encodes collaborative dynamics from local observation, which facilitates implicit coordination in realistic settings characterized by partial observability.

Abstract

Multi-Agent Coordination

Explore similar work

Distilling Collaborative Dynamics into Latent Space for Implicit Coordination in Decentralized Multi-Agent Manipulation

Multi-Agent Coordination Adaptation via Structure-Guided Orchestration

Optimization without Future Compromises? Decentralized Coordination via Collective and Reinforcement Learning