Multi-Agent Relay Thinking:MACR 如何以信心引導多智能體接力推理
Multi-Agent Relay Thinking
MACR 提出以信心引導觸發和競標機制動態轉移推理任務的多智能體接力框架,讓後續智能體承接並完善推理過程。在四個基準測試中,MACR 的準確度勝過最強單模型策略,在 MathVista 和 MMStar 分別提升 3.3% 和 3.9%;在 HallusionBench 和 MMStar 上,token 開銷比多智能體策略少 6.0 倍。
Published on Oct 27
·
on Nov 3
Authors:
,
,
,
Abstract
MACR introduces a relay-based multi-agent framework that dynamically transfers reasoning between agents using confidence-guided triggering and bidding to improve accuracy with lower token overhead.
Existing multi-agent collaboration (MAC) methods have primarily explored a group thinking collaboration topology, where agents independently generate solutions and aggregate their outputs. In this work, we investigate an alternative relay thinking topology, where reasoning ownership is dynamically transferred across agents, allowing subsequent agents to inherit and refine an evolving reasoning trajectory. We introduce Multi-Agent Confidence-guided Relay (MACR), a framework that enables dynamical relay collaboration among agents. The key challenge is determining when reasoning should be relayed and which agent should continue the process. method addresses this challenge through confidence-guided relay triggering and a bidding mechanism that selects the next agent based on its confidence. Experiments on four benchmarks show that method outperforms the strongest single-model strategies in accuracy, e.g., +3.3% on MathVista and +3.9% on MMStar, and remains on par with multi-agent strategies while consuming 6.0x less token overhead on HallusionBench and MMStar. Further analysis demonstrates that relay collaboration enables heterogeneous agents to progressively integrate complementary insights, combining heterogeneous capabilities across reasoning stages. Our source code is available at https://github.com/67L1/MACR.
View arXiv page View PDF GitHub Add to collection
Get this paper in your agent:
hf papers read 2610.35219
Don't have the latest CLI?
curl -LsSf https://hf.co/cli/install.sh | bash
Models citing this paper 0
No model linking this paper
Cite arxiv.org/abs/2610.35219 in a model README.md to link it from this page.
Datasets citing this paper 0
No dataset linking this paper
Cite arxiv.org/abs/2610.35219 in a dataset README.md to link it from this page.
Spaces citing this paper 0
No Space linking this paper
Cite arxiv.org/abs/2610.35219 in a Space README.md to link it from this page.
Collections including this paper 0
No Collection including this paper
Add this paper to a collection to link it from this page.
來源:HuggingFace Daily Papers(社區熱門論文) · huggingface.co