Auditing a Model-Hub Benchmark: Data Quality, Training Budgets, and Side Information in HuggingBench
Organizations: NASK National Research Institute Warsaw, Poland
Abstract
Benchmarks built from model hubs combine interaction data, knowledge graphs (KGs) and resource documentation, but valid files do not ensure that these artifacts describe the same objects. Treating the release as a data product, we audit HuggingBench (SIGIR 2025), a benchmark for recommending Hugging Face resources, through data validation, a replication of its eleven recommendation methods and controlled experiments with three training seeds. In three of its four external KGs, the files pass key and range checks but violate the item-entity identity constraint on which KG recommenders rely. Correcting this join improves every KG-based method in every seed, by up to 73% in mean Recall@10, and consistently reverses ten of 27 comparisons between KG configurations. The protocol matters as well: the 100-epoch budget ends before early stopping in seven to eight of 29 validation-selected settings, and 21 of 23 such setting-seed pairs improve with a longer budget. The released item text, which no reported method uses, is the strongest signal: closed-form linear models with text and the item graph exceed the best reported Recall@10 by 18% while fitting in minutes on a CPU. Checks of other recommendation releases show that individually valid releases of one dataset can use incompatible identifiers. We release corrected files, mappings, validators and logs, and derive practices for preserving identity and provenance across the artifacts of benchmarks and model lakes.
Figures & tables
| Issue | Detection | Effect | Our handling |
|---|---|---|---|
| Six encodings of the same interactions (two unnamed matrix sets, one file set per KG) | Join named file sets on names; compare degree profiles of unnamed files | Per-user results of models trained on different file sets are not comparable; item metadata cannot be attached | Name joins; permutations of the unnamed matrices by color refinement (Sec. IV-B ); release the mappings |
| Item–entity misalignment in three of four external KGs | Join items and entities by name; assert item = entity | KG side information detached from its items; KG methods underestimated by up to 73% (Sec. VI ) | Corrected KG files; alignment validator |
| 100-epoch budget truncates training | Best epoch of the selected configuration at the end of the budget | 7–8 of 29 settings exhaust the budget, depending on the seed; 21 of 23 truncated runs improve, the consistently truncated ones by up to 31% (Sec. VII ) | Longer budget with unchanged early stopping; report the best epoch and flag it for a convergence check |
| Crawl-time popularity counters may include held-out interactions | Compare counters with benchmark likes per split | Potential leakage; no gain as an additive prior for EASE (Sec. IV-D ) | Do not use; document |
| Model cards and node attributes contain the targets of the provenance tasks | Search card text and node fields for the target | A card-reading baseline nearly matches the best reported tracing method; label-naming node attributes are a potential leakage channel for classification (Sec. IX ) | Report text baselines; mask identifiers |
| KG | Triples | Item ends | Correct | Wrong item | Items | |
|---|---|---|---|---|---|---|
| 1-hop | rel. | 1,180,911 | 767,927 | 0 | 117,260 | 0 |
| corr. | 1,180,911 | 767,927 | 767,927 | 0 | 25,080 | |
| 2-hop | rel. | 3,063,081 | 767,927 | 0 | 99,312 | 0 |
| corr. | 3,063,081 | 767,927 | 767,927 | 0 | 25,080 | |
| Publisher | rel. | 25,078 | 25,078 | 4 | 39,535 | 4 |
| corr. | 25,078 | 25,078 | 25,078 | 0 | 25,078 |
| Dataset | Releases | Within | Same ID | Kept |
|---|---|---|---|---|
| Amazon-Book | KGAT–KGCL | 0 / 0 | all 24,915 | 100% |
| Amazon-Book | KGAT–KGPolicy | 0 / 0 | 32 of 24,915 | 0.7% |
| Yelp2018 | KGAT–KGCL | 0 / 0 | all 45,538 | 100% |
| Yelp2018 | KGAT–KGPolicy | 0 / 0 | 489 of 45,538 | 2.4% |
| MovieLens-1M | MKR–RippleNet | 0 / 0 | 67 of 2,347 | – |
| Reported | Reproduced | ||||
| Method | R@10 | N@10 | R@10 | N@10 | R@10 |
| LightGCN | 0.1301 | 0.1003 | 0.1301 | 0.1003 | |
| HCCF | 0.1254 | 0.0975 | 0.1275 | 0.0988 | |
| SimGCL | 0.1515 | 0.1158 | 0.1515 | 0.1158 | |
| LightGCL | 0.1558 | 0.1198 | 0.1561 | 0.1198 | |
| AutoCF | 0.1530 | 0.1174 | 0.1529 | 0.1174 | |
| Homogeneous | 1-hop | 2-hop | Publisher | ||||
|---|---|---|---|---|---|---|---|
| Method | rel. | corr. | rel. | corr. | rel. | corr. | |
| KGCL | 0.1585 | 0.1496 | 0.1591 | 0.1503 | 0.1581 | 0.1540 | 0.1604 |
| KGRec | 0.0987 | 0.0845 | 0.1457 | 0.0937 | 0.1479 | 0.0943 | 0.1204 |
| KGRec, 500 epochs | 0.0987 | 0.1107 | 0.1481 | 0.0969 | 0.1493 | 0.1054 | 0.1379 |
| KGIN (original code) | 0.1351 | 0.1169 | 0.1445 | 0.1198 | 0.1446 | 0.1152 | 0.1475 |
| Setting | Trunc. | 100 ep. | 500 ep. | R@10 [min, max] |
|---|---|---|---|---|
| LightGCN | 3 | 0.1297 | 0.1362 | [ , ] |
| DSL | 3 | 0.1433 | 0.1456 | [ , ] |
| LightGCL | 1 | 0.1401 | 0.1400 | [ , ] |
| KGIN (orig.), 1-hop rel. | 1 | 0.1169 | 0.1171 | [ , ] |
| KGRec, 1-hop rel. | 3 | 0.0845 | 0.1107 | [ , ] |
| KGRec, 1-hop corr. | 2 | 0.1457 | 0.1481 | [ , ] |
| Method | R@10 | R@20 | R@40 | N@10 | |
|---|---|---|---|---|---|
| Reported | LightGCN | 0.1301 | 0.1932 | 0.2759 | 0.1003 |
| LightGCL | 0.1558 | 0.2228 | 0.3017 | 0.1198 | |
| KGCL (homo) | 0.1578 | 0.2237 | 0.3059 | 0.1220 | |
| KGIN (homo) | 0.0096 | 0.0146 | 0.0219 | 0.0076 | |
| Reproduced | LightGCN | 0.1301 | 0.1932 | 0.2759 | 0.1003 |
| LightGCL | 0.1561 | 0.2229 | 0.3018 | 0.1198 |
| View | Similarity | Repr. | R@10 | R@10 [95% CI] |
|---|---|---|---|---|
| bge-m3 | full | TF-IDF | 0.1846 | – |
| dense | 0.1799 | [ , ] | ||
| kNN-100 | TF-IDF | 0.1844 | – | |
| dense | 0.1780 | [ , ] | ||
| Qwen3-Emb. | full | TF-IDF | 0.1847 | – |
| dense | 0.1801 | [ , ] |
| Group | Share | LGCN | EASE | +T+G | |
| Items by training likes | |||||
| 1–4 | 0.048 | 0.0059 | 0.0057 | 0.0339 | 0.0282 |
| 5–19 | 0.186 | 0.0122 | 0.0156 | 0.0407 | 0.0250 |
| 20–99 | 0.305 | 0.0128 | 0.0249 | 0.0405 | 0.0156 |
| 100 | 0.461 | 0.1579 | 0.1976 | 0.1949 | 0.0027 |
| Items by type | |||||
| Method | R@10 | N@10 | R@10 |
|---|---|---|---|
| Random | 0.0287 | 0.0152 | |
| Lineage (parents) | 0.1863 | 0.1131 | |
| Text profile: Qwen3-Embedding-0.6B | 0.1343 | 0.0854 | |
| Text profile: bge-m3 | 0.2427 | 0.1531 | |
| Text profile: TF-IDF | 0.3229 | 0.2371 | – |
| EASE + TF-IDF (closed form) | 0.3075 | 0.2261 |