πArXiv AIβ’Stalecollected in 13h
LECT: LLM OOD Detection in Text Graphs
#ood-detection#contrastive-learninglectllm
π‘SOTA OOD detection for text graphs via LLMsβkey for robust GNNs
β‘ 30-Second TL;DR
What Changed
LLMs generate dependency-aware pseudo-OOD samples
Why It Matters
Enhances robustness of graph ML models for real-world networks facing OOD data. Enables reliable deployment in dynamic environments like social or transaction graphs.
What To Do Next
Reproduce LECT experiments on arXiv:2603.20293 using your text-graph datasets.
Who should care:Researchers & Academics
Key Points
- β’LLMs generate dependency-aware pseudo-OOD samples
- β’Energy contrastive learning distinguishes IND vs OOD nodes
- β’Outperforms SOTA on citation, social, transaction graph benchmarks
- β’Addresses distribution shift in text-attributed graphs
π°
Weekly AI Recap
Read this week's curated digest of top AI events β
πRelated Updates
AI-curated news aggregator. All content rights belong to original publishers.
Original source: ArXiv AI β
This is a summary, not the original. Read the source, or get the weekly briefing.
Weekly AI briefing
One email a week. Unsubscribe anytime.