---
title: "A Living Leaderboard for Graph Machine Learning"
source: "arXiv AI/CL/LG（论文）"
source_tier: "t2"
url: "https://arxiv.org/abs/2611.03433v1"
permalink: "https://feed.astralor.com/items/itm_12da3cfd9e44a311ea227ab8"
published: "2026-11-02T16:53:26.000Z"
score: 56
channel: "ai"
tags: ["LLM","Living Leaderboard","Graph Machine Learning","Graph","October","LLM-based","Our","Amazon-ratings"]
---

# A Living Leaderboard for Graph Machine Learning

## At a glance

Graph machine learning benchmarking results are scattered across papers that use quite different sets of datasets and baseline methods, making it difficult to tell whether a reported gain reflects architectural progress…

## Key points

- Graph machine learning benchmarking results are scattered across papers that use quite different sets of datasets and baseline methods, making it difficult to tell whether a reported gain reflects architectural progress…
- We introduce a living, graph-learning-specific leaderboard that organizes reported results by dataset, putting new results into context.
- As of 15 October 2026 our LLM-based extraction pipeline extracted 27,723 eligible results across 111 datasets from 885 peer-reviewed papers.

## Full text

Graph machine learning benchmarking results are scattered across papers that use quite different sets of datasets and baseline methods, making it difficult to tell whether a reported gain reflects architectural progress or merely an outdated comparison or weakly tuned baseline. We introduce a living, graph-learning-specific leaderboard that organizes reported results by dataset, putting new results into context. As of 15 October 2026 our LLM-based extraction pipeline extracted 27,723 eligible results across 111 datasets from 885 peer-reviewed papers. Our analysis of these results quantifies concerns about graph-learning benchmarking: legacy datasets are still actively used and many result tables contain baseline results that are drastically lower than those already available in the literature, e.g. 45% Amazon-ratings tables and 37% for Peptides-struct. The leaderboard allows to quickly compare new results against compatible prior results, aiming to improve the overall quality of graph learning benchmarking.

---

Verify against the original before citing; curated by Astral Feed: https://feed.astralor.com/items/itm_12da3cfd9e44a311ea227ab8