Remove unnecessary FineWeb overlap section

README.md

@@ -89,20 +89,6 @@ The heuristic rules are organized into four groups:
 
 Deduplication is performed within each Common Crawl dump rather than across the full collection.
 
-## 📈 Comparison with FineWeb
-
-For a URL-level comparison on `CC-MAIN-2025-26`, the deduplicated Ultra-FineWeb-L1 output shared 175,281,630 URLs with FineWeb. This corresponds to 90.61% recall and 90.16% precision for this validation run. The comparison is taken immediately after MinHash deduplication, before the additional sanitation pass, so both pipelines are measured at equivalent stages.
-
-| Metric | Before MinHash | After MinHash |
-|:---|---:|---:|
-| Ultra-FineWeb-L1 unique URLs | 258,071,345 | 194,402,835 |
-| FineWeb unique URLs | 193,438,138 | 193,438,138 |
-| URL intersection | 182,421,184 | 175,281,630 |
-| Recall against FineWeb | 94.30% | **90.61%** |
-| Precision | 70.69% | **90.16%** |
-
-On this validation dump, MinHash deduplication removes many URLs that are not present in FineWeb: precision increases from 70.69% to 90.16%, while recall decreases from 94.30% to 90.61%. This reflects a trade-off in URL coverage and deduplication, not a direct measurement of overall text quality. The figures describe URL overlap for this dump only; they do not imply that the two datasets contain identical documents or that the same metrics apply to all released dumps.
-
 ## 📁 Dataset Structure
 
 Ultra-FineWeb-L1 is organized by Common Crawl dump: