CoDe: Communication Delay-Tolerant Multi-Agent Collaboration via Dual Alignment of Intent and Timeliness

Shoucheng Song, Youfang Lin, Sheng Han, Chang Yao, Hao Wu, Shuo Wang, Kai Lv · 2025

Read CoDe: Communication Delay-Tolerant Multi-Agent Collaboration via Dual Alignment of Intent and Timeliness by Shoucheng Song, Youfang Lin, Sheng Han, Chang Yao, Hao Wu, Shuo Wang, Kai Lv (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2501-05207.txt.

Description

Song and co-authors help robot teams cope with late messages. Multi-agent systems coordinate by talking, but real channels delay messages, so agents act on stale news about each other and teamwork frays. Their framework, CoDe, attacks the problem twice. First, each message carries an intent representation, inferred from the sender's likely future actions, which stays meaningful even when it arrives late.

Second, a dual alignment step weighs messages by both intent and timeliness, favoring recent ones that match what the agent needs. On three multi-agent benchmarks, CoDe beats the baselines without delay and stays robust when delays are fixed or shifting. The authors present delayed communication as a solvable design problem rather than a fact of life.

Catalog record

CREATORShoucheng Song, Youfang Lin, Sheng Han, Chang Yao, Hao Wu, Shuo Wang, Kai Lv
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE32 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2501-05207

Browse the full library · About the Tower of Babel