# Factual Consistency Evaluation on SAMSum

Source: [Factual Consistency Evaluation on SAMSum](<https://www.sota2.com/research/sota/factual-consistency-evaluation-on-samsum>)

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

Data updated: 2026-02-26T06:04:34.723Z

Evaluation of model performance on Factual Consistency using the SAMSum dataset, measured via Spearman correlation.

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 | Factual Consistency Evaluation |
| Dataset (including any stated split) | SAMSum |
| Metric | Spearman Correlation |
| Unit | SCALAR |
| Direction | Higher is better |
| Results loaded | 30 of 30 |
| 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

- [Spearman Correlation](<https://www.sota2.com/research/sota/factual-consistency-evaluation-on-samsum.md?metric=Spearman%20Correlation>): Higher is better; unit: SCALAR.

## Recorded results

| Method | Spearman Correlation | Result date | Report paper |
| --- | --- | --- | --- |
| ALIGN | 46.7 | Not recorded | 68fa71131e209131cea4f1ee |
| ALIGNSCORE-large | 43.8 | Not recorded | 68fa6f801e209131cea4607f |
| QAFactEval | 35.9 | Not recorded | 68fa6f801e209131cea4607f |
| QAFactEval | 35.9 | Not recorded | 68fa71131e209131cea4f1ee |
| MNLI | 31.1 | Not recorded | 68fa6f801e209131cea4607f |
| ALIGNSCORE-base | 23.4 | Not recorded | 68fa6f801e209131cea4607f |
| UniEval | 22.8 | Not recorded | 68fa6f801e209131cea4607f |
| UniEval | 22.8 | Not recorded | 68fa71131e209131cea4f1ee |
| ALIGN | 22.3 | Not recorded | 68fa71131e209131cea4f1ee |
| DAE | 18.6 | Not recorded | 68fa6f801e209131cea4607f |
| CTC | 17.7 | Not recorded | 68fa6f801e209131cea4607f |
| CTC | 17.7 | Not recorded | 68fa71131e209131cea4f1ee |
| ROUGE-2 | 17.5 | Not recorded | 68fa6f801e209131cea4607f |
| NER-Overlap | 16.7 | Not recorded | 68fa6f801e209131cea4607f |
| ROUGE-L | 16.3 | Not recorded | 68fa6f801e209131cea4607f |
| BARTScore | 16.3 | Not recorded | 68fa6f801e209131cea4607f |
| BARTScore | 16.3 | Not recorded | 68fa71131e209131cea4f1ee |
| ROUGE-1 | 15.1 | Not recorded | 68fa6f801e209131cea4607f |
| SummaC-ZS | 15.1 | Not recorded | 68fa6f801e209131cea4607f |
| BLEU | 13.7 | Not recorded | 68fa6f801e209131cea4607f |
| BERTScore | 13.1 | Not recorded | 68fa6f801e209131cea4607f |
| BERTScore | 13.1 | Not recorded | 68fa71131e209131cea4f1ee |
| SummaC-CONV | 9.8 | Not recorded | 68fa6f801e209131cea4607f |
| BLANC | 9.1 | Not recorded | 68fa6f801e209131cea4607f |
| SimCSE | 7.9 | Not recorded | 68fa6f801e209131cea4607f |
| BLEURT | 6.7 | Not recorded | 68fa6f801e209131cea4607f |
| BLEURT | 6.7 | Not recorded | 68fa71131e209131cea4f1ee |
| QuestEval | 3.9 | Not recorded | 68fa6f801e209131cea4607f |
| FEQA | 0 | Not recorded | 68fa6f801e209131cea4607f |
| FactCC | -4.4 | Not recorded | 68fa6f801e209131cea4607f |

### ALIGN

Spearman Correlation: 46.7. Result ID: 699197373a6c314849093104.

[Reporting paper](<https://www.sota2.com/research/paper/68fa71131e209131cea4f1ee.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa71131e209131cea4f1ee.md>)

[Implementation](<https://github.com/yuh-zha/Align>)

Recorded evaluation configuration: {"Model size":"large"}. This record does not assert independent reproduction.

### ALIGNSCORE-large

Spearman Correlation: 43.8. Result ID: 6990c73d3a6c3148490335bb.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6f801e209131cea4607f.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6f801e209131cea4607f.md>)

[Implementation](<https://github.com/yuh-zha/AlignScore>)

Recorded evaluation configuration: {"Type":"Ours"}. This record does not assert independent reproduction.

### QAFactEval

Spearman Correlation: 35.9. Result ID: 6990c73d3a6c3148490335a8.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6f801e209131cea4607f.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6c611e209131cea33b79.md>)

Recorded evaluation configuration: {"Type":"QA"}. This record does not assert independent reproduction.

### QAFactEval

Spearman Correlation: 35.9. Result ID: 699197373a6c314849093102.

[Reporting paper](<https://www.sota2.com/research/paper/68fa71131e209131cea4f1ee.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6c611e209131cea33b79.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### MNLI

Spearman Correlation: 31.1. Result ID: 6990c73d3a6c3148490335b1.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6f801e209131cea4607f.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa64101e209131cea02761.md>)

Recorded evaluation configuration: {"Type":"NLI"}. This record does not assert independent reproduction.

### ALIGNSCORE-base

Spearman Correlation: 23.4. Result ID: 6990c73d3a6c3148490335ba.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6f801e209131cea4607f.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6f801e209131cea4607f.md>)

[Implementation](<https://github.com/yuh-zha/AlignScore>)

Recorded evaluation configuration: {"Type":"Ours"}. This record does not assert independent reproduction.

### UniEval

Spearman Correlation: 22.8. Result ID: 6990c73d3a6c3148490335b5.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6f801e209131cea4607f.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6da11e209131cea3afdb.md>)

Recorded evaluation configuration: {"Type":"Misc"}. This record does not assert independent reproduction.

### UniEval

Spearman Correlation: 22.8. Result ID: 699197373a6c314849093101.

[Reporting paper](<https://www.sota2.com/research/paper/68fa71131e209131cea4f1ee.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6da11e209131cea3afdb.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### ALIGN

Spearman Correlation: 22.3. Result ID: 699197373a6c314849093103.

[Reporting paper](<https://www.sota2.com/research/paper/68fa71131e209131cea4f1ee.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa71131e209131cea4f1ee.md>)

[Implementation](<https://github.com/yuh-zha/Align>)

Recorded evaluation configuration: {"Model size":"base"}. This record does not assert independent reproduction.

### DAE

Spearman Correlation: 18.6. Result ID: 6990c73d3a6c3148490335b2.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6f801e209131cea4607f.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa68891e209131cea1cc4b.md>)

Recorded evaluation configuration: {"Type":"NLI"}. This record does not assert independent reproduction.

### CTC

Spearman Correlation: 17.7. Result ID: 6990c73d3a6c3148490335b6.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6f801e209131cea4607f.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6ba71e209131cea2f996.md>)

Recorded evaluation configuration: {"Type":"Misc"}. This record does not assert independent reproduction.

### CTC

Spearman Correlation: 17.7. Result ID: 699197373a6c314849093100.

[Reporting paper](<https://www.sota2.com/research/paper/68fa71131e209131cea4f1ee.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6ba71e209131cea2f996.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### ROUGE-2

Spearman Correlation: 17.5. Result ID: 6990c73d3a6c3148490335aa.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6f801e209131cea4607f.md>)

Recorded evaluation configuration: {"Type":"Similarity Matching"}. This record does not assert independent reproduction.

### NER-Overlap

Spearman Correlation: 16.7. Result ID: 6990c73d3a6c3148490335ae.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6f801e209131cea4607f.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6b421e209131cea2d015.md>)

Recorded evaluation configuration: {"Type":"Similarity Matching"}. This record does not assert independent reproduction.

### ROUGE-L

Spearman Correlation: 16.3. Result ID: 6990c73d3a6c3148490335ab.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6f801e209131cea4607f.md>)

Recorded evaluation configuration: {"Type":"Similarity Matching"}. This record does not assert independent reproduction.

### BARTScore

Spearman Correlation: 16.3. Result ID: 6990c73d3a6c3148490335b7.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6f801e209131cea4607f.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6b161e209131cea2c0f5.md>)

[Implementation](<https://github.com/neulab/BARTScore>)

Recorded evaluation configuration: {"Type":"Misc"}. This record does not assert independent reproduction.

### BARTScore

Spearman Correlation: 16.3. Result ID: 699197373a6c3148490930ff.

[Reporting paper](<https://www.sota2.com/research/paper/68fa71131e209131cea4f1ee.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6b161e209131cea2c0f5.md>)

[Implementation](<https://github.com/neulab/BARTScore>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### ROUGE-1

Spearman Correlation: 15.1. Result ID: 6990c73d3a6c3148490335a9.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6f801e209131cea4607f.md>)

Recorded evaluation configuration: {"Type":"Similarity Matching"}. This record does not assert independent reproduction.

### SummaC-ZS

Spearman Correlation: 15.1. Result ID: 6990c73d3a6c3148490335b3.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6f801e209131cea4607f.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6b421e209131cea2d015.md>)

Recorded evaluation configuration: {"Type":"NLI"}. This record does not assert independent reproduction.

### BLEU

Spearman Correlation: 13.7. Result ID: 6990c73d3a6c3148490335ac.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6f801e209131cea4607f.md>)

Recorded evaluation configuration: {"Type":"Similarity Matching"}. This record does not assert independent reproduction.

### BERTScore

Spearman Correlation: 13.1. Result ID: 6990c73d3a6c3148490335ad.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6f801e209131cea4607f.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa67171e209131cea140d5.md>)

Recorded evaluation configuration: {"Type":"Similarity Matching"}. This record does not assert independent reproduction.

### BERTScore

Spearman Correlation: 13.1. Result ID: 699197373a6c3148490930fd.

[Reporting paper](<https://www.sota2.com/research/paper/68fa71131e209131cea4f1ee.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa67171e209131cea140d5.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### SummaC-CONV

Spearman Correlation: 9.8. Result ID: 6990c73d3a6c3148490335b4.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6f801e209131cea4607f.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6b421e209131cea2d015.md>)

Recorded evaluation configuration: {"Type":"NLI"}. This record does not assert independent reproduction.

### BLANC

Spearman Correlation: 9.1. Result ID: 6990c73d3a6c3148490335b9.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6f801e209131cea4607f.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa68b71e209131cea1db13.md>)

Recorded evaluation configuration: {"Type":"Misc"}. This record does not assert independent reproduction.

### SimCSE

Spearman Correlation: 7.9. Result ID: 6990c73d3a6c3148490335af.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6f801e209131cea4607f.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6c9c1e209131cea34e2c.md>)

[Implementation](<https://github.com/princeton-nlp/SimCSE>)

Recorded evaluation configuration: {"Type":"Similarity Matching"}. This record does not assert independent reproduction.

### BLEURT

Spearman Correlation: 6.7. Result ID: 6990c73d3a6c3148490335b0.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6f801e209131cea4607f.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa67961e209131cea173db.md>)

Recorded evaluation configuration: {"Type":"Regression"}. This record does not assert independent reproduction.

### BLEURT

Spearman Correlation: 6.7. Result ID: 699197373a6c3148490930fe.

[Reporting paper](<https://www.sota2.com/research/paper/68fa71131e209131cea4f1ee.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa67961e209131cea173db.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### QuestEval

Spearman Correlation: 3.9. Result ID: 6990c73d3a6c3148490335a7.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6f801e209131cea4607f.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa69b91e209131cea23bee.md>)

Recorded evaluation configuration: {"Type":"QA"}. This record does not assert independent reproduction.

### FEQA

Spearman Correlation: 0. Result ID: 6990c73d3a6c3148490335a6.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6f801e209131cea4607f.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa68841e209131cea1c9f0.md>)

Recorded evaluation configuration: {"Type":"QA"}. This record does not assert independent reproduction.

### FactCC

Spearman Correlation: -4.4. Result ID: 6990c73d3a6c3148490335b8.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6f801e209131cea4607f.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa66911e209131cea10f01.md>)

Recorded evaluation configuration: {"Type":"Misc"}. This record does not assert independent reproduction.

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