.png){kind=icon}
.png){kind=icon}
Return
A Scalable Graph Neural Framework for Cross-Platform Trust and Intent Modeling in Large Language Model Systems
.png){kind=icon}
PRE
AI
A
S
A
K
DOI：10.1016/j.compeleceng.2026.111332.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
• A cross-platform trust and intent alignment framework for LLM evaluation is proposed. • Real-world conversations from five LLM platforms are analyzed for trust signals. • Graph-based dialogue modeling captures trust evolution across multi-turn exchanges. • Platform-specific features improve intent satisfaction prediction by 5.7 F1 points. • Experiments span 142,808 conversations in 101 languages across five platforms.
Journal
C
IF:
4.9
Papers:
6.7K
Citations:
1.3W
Organization
Cited Papers