Clustering on OpenAI Codex (test)
80AccuracyCLUSTERFUSION
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| CLUSTERFUSIONBackbone=GPT-4o, Order/Setting=Assignment only2025.12 | 80 | 81.6 | |
| CLUSTERFUSIONBackbone=GPT-4o, Order/Setting=Oracle order2025.12 | 69.9 | 75.4 | |
| CLUSTERFUSIONBackbone=GPT-4o, Order/Setting=Cosine order2025.12 | 66 | 72.6 | |
| CLUSTERFUSIONBackbone=GPT-4o, Order/Setting=KMeans order2025.12 | 63.6 | 71.5 | |
| CLUSTERFUSIONBackbone=GPT-4o, Order/Setting=Unsorted2025.12 | 61.8 | 70.5 | |
| Agglomerative Clust.2025.12 | 44.5 | 56.1 | |
| KMeans2025.12 | 41.7 | 55.5 | |
| Keyphrase Clust. (w/ KMeans)Backbone=GPT-4o2025.12 | 40.3 | 52.7 | |
| Keyphrase Clust. (w/ Agglom.)Backbone=GPT-4o2025.12 | 35.8 | 48.9 |