# Overlap ratio estimation on Wiki (held-out pairs)

Source: [Overlap ratio estimation on Wiki (held-out pairs)](<https://www.sota2.com/research/sota/overlap-ratio-estimation-on-wiki-held-out-pairs>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-07-07T08:10:32.363Z

Evaluation of Overlap ratio estimation performance on the held-out pairs split of the Wiki dataset, measured using MAE.

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

| Scope | Recorded value |
| --- | --- |
| Task | Overlap ratio estimation |
| Dataset (including any stated split) | Wiki (held-out pairs) |
| Metric | MAE |
| Unit | SCALAR |
| Direction | Lower is better |
| Results loaded | 13 of 13 |
| Coverage | All results currently returned for this benchmark |

Evaluation settings may differ between results. The original papers and recorded configurations are needed to establish comparability. A shared metric name alone does not establish an identical protocol.

## Metrics

- [MAE](<https://www.sota2.com/research/sota/overlap-ratio-estimation-on-wiki-held-out-pairs.md?metric=MAE>): Lower is better; unit: SCALAR.

## Recorded results

| Method | MAE | Result date | Report paper |
| --- | --- | --- | --- |
| ALORE | 0.04 | Not recorded | 6a4c87e085dba69fef8b237b |
| Armadillo | 0.069 | Not recorded | 6a4c87e085dba69fef8b237b |
| BERT-T | 0.097 | Not recorded | 6a4c87e085dba69fef8b237b |
| RoBERTa-T | 0.101 | Not recorded | 6a4c87e085dba69fef8b237b |
| BERT-R | 0.115 | Not recorded | 6a4c87e085dba69fef8b237b |
| RoBERTa-R | 0.135 | Not recorded | 6a4c87e085dba69fef8b237b |
| TURL | 0.199 | Not recorded | 6a4c87e085dba69fef8b237b |
| BERT-HT | 0.207 | Not recorded | 6a4c87e085dba69fef8b237b |
| Jaccard | 0.223 | Not recorded | 6a4c87e085dba69fef8b237b |
| RoBERTa-HT | 0.229 | Not recorded | 6a4c87e085dba69fef8b237b |
| Jaccard-BU | 0.231 | Not recorded | 6a4c87e085dba69fef8b237b |
| Jaccard-BN | 0.278 | Not recorded | 6a4c87e085dba69fef8b237b |
| EmbDI | 0.372 | Not recorded | 6a4c87e085dba69fef8b237b |

### ALORE

MAE: 0.04. Result ID: 6a4cb45d60b88bb339c8346a.

[Reporting paper](<https://www.sota2.com/research/paper/6a4c87e085dba69fef8b237b.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a4c87e085dba69fef8b237b.md>)

Recorded evaluation configuration: {"approach":"alignment-guided overlap ratio estimator"}. This record does not assert independent reproduction.

### Armadillo

MAE: 0.069. Result ID: 6a4cb45d60b88bb339c83469.

[Reporting paper](<https://www.sota2.com/research/paper/6a4c87e085dba69fef8b237b.md>)

Recorded evaluation configuration: {"approach":"tripartite graph embedding with GraphSAGE"}. This record does not assert independent reproduction.

### BERT-T

MAE: 0.097. Result ID: 6a4cb45d60b88bb339c83462.

[Reporting paper](<https://www.sota2.com/research/paper/6a4c87e085dba69fef8b237b.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa65c31e209131cea0c4ed.md>)

[Implementation](<https://github.com/google-research/bert>)

Recorded evaluation configuration: {"serialization":"Table (-T)"}. This record does not assert independent reproduction.

### RoBERTa-T

MAE: 0.101. Result ID: 6a4cb45d60b88bb339c83465.

[Reporting paper](<https://www.sota2.com/research/paper/6a4c87e085dba69fef8b237b.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa66131e209131cea0e18b.md>)

[Implementation](<https://github.com/pytorch/fairseq>)

Recorded evaluation configuration: {"serialization":"Table (-T)"}. This record does not assert independent reproduction.

### BERT-R

MAE: 0.115. Result ID: 6a4cb45d60b88bb339c83461.

[Reporting paper](<https://www.sota2.com/research/paper/6a4c87e085dba69fef8b237b.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa65c31e209131cea0c4ed.md>)

[Implementation](<https://github.com/google-research/bert>)

Recorded evaluation configuration: {"serialization":"Row (-R)"}. This record does not assert independent reproduction.

### RoBERTa-R

MAE: 0.135. Result ID: 6a4cb45d60b88bb339c83464.

[Reporting paper](<https://www.sota2.com/research/paper/6a4c87e085dba69fef8b237b.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa66131e209131cea0e18b.md>)

[Implementation](<https://github.com/pytorch/fairseq>)

Recorded evaluation configuration: {"serialization":"Row (-R)"}. This record does not assert independent reproduction.

### TURL

MAE: 0.199. Result ID: 6a4cb45d60b88bb339c83467.

[Reporting paper](<https://www.sota2.com/research/paper/6a4c87e085dba69fef8b237b.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa68d51e209131cea1e7dc.md>)

[Implementation](<https://github.com/sunlab-osu/TURL>)

Recorded evaluation configuration: {"approach":"contextualized cell representations"}. This record does not assert independent reproduction.

### BERT-HT

MAE: 0.207. Result ID: 6a4cb45d60b88bb339c83463.

[Reporting paper](<https://www.sota2.com/research/paper/6a4c87e085dba69fef8b237b.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa65c31e209131cea0c4ed.md>)

[Implementation](<https://github.com/google-research/bert>)

Recorded evaluation configuration: {"serialization":"Hashed-Table (-HT)"}. This record does not assert independent reproduction.

### Jaccard

MAE: 0.223. Result ID: 6a4cb45d60b88bb339c8345e.

[Reporting paper](<https://www.sota2.com/research/paper/6a4c87e085dba69fef8b237b.md>)

Recorded evaluation configuration: {"variant":"set-based Jaccard"}. This record does not assert independent reproduction.

### RoBERTa-HT

MAE: 0.229. Result ID: 6a4cb45d60b88bb339c83466.

[Reporting paper](<https://www.sota2.com/research/paper/6a4c87e085dba69fef8b237b.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa66131e209131cea0e18b.md>)

[Implementation](<https://github.com/pytorch/fairseq>)

Recorded evaluation configuration: {"serialization":"Hashed-Table (-HT)"}. This record does not assert independent reproduction.

### Jaccard-BU

MAE: 0.231. Result ID: 6a4cb45d60b88bb339c8345f.

[Reporting paper](<https://www.sota2.com/research/paper/6a4c87e085dba69fef8b237b.md>)

Recorded evaluation configuration: {"variant":"bag-union variant (-BU)"}. This record does not assert independent reproduction.

### Jaccard-BN

MAE: 0.278. Result ID: 6a4cb45d60b88bb339c83460.

[Reporting paper](<https://www.sota2.com/research/paper/6a4c87e085dba69fef8b237b.md>)

Recorded evaluation configuration: {"variant":"bag-normalized variant (-BN)"}. This record does not assert independent reproduction.

### EmbDI

MAE: 0.372. Result ID: 6a4cb45d60b88bb339c83468.

[Reporting paper](<https://www.sota2.com/research/paper/6a4c87e085dba69fef8b237b.md>)

Recorded evaluation configuration: {"approach":"tripartite graph node2vec adaptation"}. This record does not assert independent reproduction.

Page 1 of 1. This is an index page, not the full collection.
