# Identifying Group Anchors in Real-World Group Interactions Under Label Scarcity

arXiv: https://arxiv.org/abs/2509.20762 · HTML: https://mlsanity.com/paper/2509.20762/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 28 of its 67 references that are in the map.
- First posted on arXiv 25 Sep 2025.

## Builds on (most cited first, 28 of 28)

- [Adam: A Method for Stochastic Optimization](https://mlsanity.com/paper/1412.6980/) (2014)
- [PyTorch: An Imperative Style, High-Performance Deep Learning Library](https://mlsanity.com/paper/1912.01703/) (2019)
- [Hypergraph Neural Networks](https://mlsanity.com/paper/1809.09401/) (2018)
- [Hypergraph Convolution and Hypergraph Attention](https://mlsanity.com/paper/1901.08150/) (2019)
- [UniGNN: a Unified Framework for Graph and Hypergraph Neural Networks](https://mlsanity.com/paper/2105.00956/) (2021)
- [HNHN: Hypergraph Networks with Hyperedge Neurons](https://mlsanity.com/paper/2006.12278/) (2020)
- [Motifs in Temporal Networks](https://mlsanity.com/paper/1612.09259/) (2016)
- [Simplicial Closure and higher-order link prediction](https://mlsanity.com/paper/1802.06916/) (2018)
- [Multilayer Networks](https://mlsanity.com/paper/1309.7233/) (2013)
- [The structure and dynamics of multilayer networks](https://mlsanity.com/paper/1407.0742/) (2014)
- [Analyzing Overfitting under Class Imbalance in Neural Networks for Image Segmentation](https://mlsanity.com/paper/2102.10365/) (2021)
- [How Much and When Do We Need Higher-order Information in Hypergraphs? A Case Study on Hyperedge Prediction](https://mlsanity.com/paper/2001.11181/) (2020)
- [A Survey on Hypergraph Neural Networks: An In-Depth and Step-By-Step Guide](https://mlsanity.com/paper/2404.01039/) (2024)
- [Semi-supervised Hypergraph Node Classification on Hypergraph Line Expansion](https://mlsanity.com/paper/2005.04843/) (2020)
- [Classification of Edge-dependent Labels of Nodes in Hypergraphs](https://mlsanity.com/paper/2306.03032/) (2023)
- [AHP: Learning to Negative Sample for Hyperedge Prediction](https://mlsanity.com/paper/2204.06353/) (2022)
- [Three hypergraph eigenvector centralities](https://mlsanity.com/paper/1807.09644/) (2018)
- [A Hypergraph Neural Network Framework for Learning Hyperedge-Dependent Node Embeddings](https://mlsanity.com/paper/2212.14077/) (2022)
- [Node and Edge Nonlinear Eigenvector Centrality for Hypergraphs](https://mlsanity.com/paper/2101.06215/) (2021)
- [Core-periphery detection in hypergraphs](https://mlsanity.com/paper/2202.12769/) (2022)
- [Encapsulation Structure and Dynamics in Hypergraphs](https://mlsanity.com/paper/2307.04613/) (2023)
- [Planted Hitting Set Recovery in Hypergraphs](https://mlsanity.com/paper/1905.05839/) (2019)
- [Annotated Hypergraphs: Models and Applications](https://mlsanity.com/paper/1911.01331/) (2019)
- [Core-periphery Models for Hypergraphs](https://mlsanity.com/paper/2206.00783/) (2022)
- [Hcore-Init: Neural Network Initialization based on Graph Degeneracy](https://mlsanity.com/paper/2004.07636/) (2020)
- [Hypercore Decomposition for Non-Fragile Hyperedges: Concepts, Algorithms, Observations, and Applications](https://mlsanity.com/paper/2301.08440/) (2023)
- [Neighborhood-based Hypergraph Core Decomposition](https://mlsanity.com/paper/2301.06426/) (2023)
- [Exploring Cohesive Subgraphs in Hypergraphs: The (k,g)-core Approach](https://mlsanity.com/paper/2309.04350/) (2023)

## Abstract (arXiv)

Group interactions occur in various real-world contexts, e.g., co-authorship, email communication, and online Q&A. In each group, there is often a particularly significant member, around whom the group is formed. Examples include the first or last author of a paper, the sender of an email, and the questioner in a Q&A session. In this work, we discuss the existence of such individuals in real-world group interactions. We call such individuals group anchors and study the problem of identifying them. First, we introduce the concept of group anchors and the identification problem. Then, we discuss our observations on group anchors in real-world group interactions. Based on our observations, we develop AnchorRadar, a fast and effective method for group anchor identification under realistic settings with label scarcity, i.e., when only a few groups have known anchors. AnchorRadar is a semi-supervised method using information from groups both with and without known group anchors. Finally, through extensive experiments on thirteen real-world datasets, we demonstrate the empirical superiority of AnchorRadar over various baselines w.r.t. accuracy and efficiency. In most cases, AnchorRadar achieves higher accuracy in group anchor identification than all the baselines, while using 10.2× less training time than the fastest baseline and 43.6× fewer learnable parameters than the most lightweight baseline on average.