ControBench: An Interaction-Aware Benchmark for Controversial Discourse Analysis on Social Networks
Not on the map. Why?
Abstract (arXiv)
Understanding how people argue across ideological divides online is important for studying political polarization, misinformation, and content moderation. Existing datasets capture only part of this problem: some preserve text but ignore interaction structure, some model structure without rich semantics, and others represent conversations without stable user-level ideological identity. We introduce ControBench, a benchmark for controversial discourse analysis that combines heterogeneous social interaction graphs with rich textual semantics. Built from Reddit discussions on three topics, Trump, abortion, and religion, ControBench contains 7,370 users, 1,783 posts, and 26,525 interactions. The graph contains user and post nodes connected by semantically enriched edges; in particular, user-comment-user edges encode both a reply and the parent comment that it responds to, preserving local argumentative context. User labels are derived from self-declared Reddit flairs, providing a scalable proxy for ideological identity without manual annotation. The resulting datasets exhibit low or negative adjusted homophily (Trump: -0.77, Abortion: 0.06, Religion: 0.04), reflecting the cross-cutting structure of real-world debate. We evaluate graph neural networks, pretrained language models, and large language models on ControBench and observe distinct performance patterns across topics and model families, especially when ideological boundaries are ambiguous. These results position ControBench as a challenging and realistic benchmark for controversial discourse analysis.
Key facts. No citing papers in the map yet (data through 7 Oct 2026). Builds on 22 of its 43 references that are in the map. First posted on arXiv 1 May 2026.
Builds on
22 of 43 references.
References · 22
- 2018BERT: Pre-training of Deep Bidirectional Transformers for Language Understandingcited by 40,204
- 2019RoBERTa: A Robustly Optimized BERT Pretraining Approachcited by 13,266
- 2019Sentence-BERT: Sentence Embeddings using Siamese BERT-Networkscited by 9,049
- 2017Inductive Representation Learning on Large Graphscited by 6,284
- 2021SimCSE: Simple Contrastive Learning of Sentence Embeddingscited by 2,168
- 2017Modeling Relational Data with Graph Convolutional Networkscited by 1,684
- 2019Heterogeneous Graph Attention Networkcited by 732
- 2022Revisiting Heterophily For Graph Neural Networkscited by 169
- 2022Learning on Large-scale Text-attributed Graphs via Variational Inferencecited by 153
- 2022Characterizing Graph Datasets for Node Classification: Homophily-Heterophily Dichotomy and Beyondcited by 60
- 2023Stance Detection with Collaborative Role-Infused LLM-Based Agentscited by 53
- 2021Twitch Gamers: a Dataset for Evaluating Proximity Preserving and Structural Role-based Node Embeddingscited by 48
10 more
- 2024The Heterophilic Graph Learning Handbook: Benchmarks, Models, Theoretical Analysis, Applications and Challengescited by 29
- 2024DTGB: A Comprehensive Benchmark for Dynamic Text-Attributed Graphscited by 26
- 2024A Challenge Dataset and Effective Models for Conversational Stance Detectioncited by 17
- 2024The Power of LLM-Generated Synthetic Data for Stance Detection in Online Political Discussionscited by 16
- 2025Zero-Shot Conversational Stance Detection: Dataset and Approachescited by 11
- 2024Exploring the Potential of Large Language Models for Heterophilic Graphscited by 10
- 2024TEG-DB: A Comprehensive Dataset and Benchmark of Textual-Edge Graphscited by 9
- 2024When Heterophily Meets Heterogeneity: Challenges and a New Large-Scale Graph Benchmarkcited by 6
- 2025HeTGB: A Comprehensive Benchmark for Heterophilic Text-Attributed Graphscited by 5
- 2025FanChuan: A Multilingual and Graph-Structured Benchmark For Parody Detection and Analysiscited by 4
Built on by
Not cited yet.