Delving into Instance-Dependent Label Noise in Graph Data: A Comprehensive Study and Benchmark Conference: Proceedings of the 31st ACM SIGKDD Conference on Knowledge Discovery and Data Mining V.2; August 3–7, 2025; Toronto, ON, Canada Proceedings of the 31st ACM SIGKDD Conference on Knowledge Discov
Suyeon Kim, SeongKu Kang, Dongwoo Kim, Jungseul Ok, Hwanjo Yu · 2025
Read Delving into Instance-Dependent Label Noise in Graph Data: A Comprehensive Study and Benchmark Conference: Proceedings of the 31st ACM SIGKDD Conference on Knowledge Discovery and Data Mining V.2; August 3–7, 2025; Toronto, ON, Canada Proceedings of the 31st ACM SIGKDD Conference on Knowledge Discov by Suyeon Kim, SeongKu Kang, Dongwoo Kim, Jungseul Ok, Hwanjo Yu (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-2506-12468.txt.
Description
Suyeon Kim and four co-authors study wrong labels in graph data in a 2025 paper. Graph networks sort nodes, such as users or papers, by the links among them, and they top most rival methods when labels are clean. Real labels are often wrong, though. Past work mostly assumed the error rate depends on the class alone. In real life, the chance of a wrong label also depends on each case by itself.
The team built a test set called BeGIN to capture that harder kind of noise. It adds noise by rules and by language model fakes, then grades many fix methods across network types. The faked noise proved the hardest to beat. Methods that tune a setting per node did best, which hints at where future work should aim. All code is shared for reuse.
Catalog record
| CREATOR | Suyeon Kim, SeongKu Kang, Dongwoo Kim, Jungseul Ok, Hwanjo Yu |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 64 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2506-12468