RQPool: A Novel Multi-Branch Graph-Level Anomaly Detection
Why this work is in the frame
A frame that forgets how it found something cannot be audited. These are the routes that admitted this work.
Bibliographic record
Abstract
Anomaly Detection (AD) is crucial across various domains,as it identifies irregularities or unusual patternsthat, if quickly addressed, can prevent financial and datalosses, protect health, and prevent disasters. Many systemssuch as social networks, communication systems,and biological networks are naturally represented asgraphs with entities as nodes and interactions as edges.By analyzing these structures, we can uncover anomaliesthat are not apparent using traditional methods.However, current Graph-based AD techniques face significantchallenges, particularly low accuracy on largerdatasets. As datasets grow larger, the complexity ofthe graphs increases. This complexity makes it morechallenging for models to distinguish normal variationsfrom true anomalies. Moreover, existing Graph NeuralNetwork (GNN) algorithms focus primarily on spatialdomain features while neglecting spectral properties.Furthermore, most algorithms concentrate on intra-graphproperties such as edges and nodes, while overlookingrich global inter-graph relationships like GraphSimilarity Measures and Cross-Graph Connectivity. Toaddress these challenges, we propose a novel hybridmethod, RQPool, which integrates intra-graph spectralproperties and inter-graph spatial properties intoa unified Graph-level Anomaly Detection classifier. Inempirical evaluations across multiple datasets, RQPoolconsistently achieves higher AUC and macro-F1 scores compared topurely spectral or spatial baselines, the current state-of-the-art approaches, particularly excelling on large-scalegraphs.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.002 | 0.001 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.002 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.004 | 0.001 |
| Research integrity | 0.000 | 0.001 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it