# Fact Verification on FEVER-Symmetric

Source: [Fact Verification on FEVER-Symmetric](<https://www.sota2.com/research/sota/fact-verification-on-fever-symmetric>)

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

Data updated: 2026-02-26T05:58:18.146Z

Evaluation of model performance on the FEVER-Symmetric dataset for Fact Verification, reporting Precision, Recall, and F1 scores.

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 | Fact Verification |
| Dataset (including any stated split) | FEVER-Symmetric |
| Metric | Precision |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 16 of 16 |
| 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

- [Precision](<https://www.sota2.com/research/sota/fact-verification-on-fever-symmetric.md?metric=Precision>): Higher is better; unit: PERCENT.

- [Recall](<https://www.sota2.com/research/sota/fact-verification-on-fever-symmetric.md?metric=Recall>): Higher is better; unit: PERCENT.

- [F1 Score](<https://www.sota2.com/research/sota/fact-verification-on-fever-symmetric.md?metric=F1%20Score>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Precision | Result date | Report paper |
| --- | --- | --- | --- |
| RoBERTa-large | 88 | Not recorded | 68fa6a0d1e209131cea25cd8 |
| BERT-base | 85.7 | Not recorded | 68fa6a0d1e209131cea25cd8 |
| RoBERTa-large | 85.5 | Not recorded | 68fa6a0d1e209131cea25cd8 |
| BERT-base | 81.5 | Not recorded | 68fa6a0d1e209131cea25cd8 |
| QACG (RoBERTa-large) | 77.3 | Not recorded | 68fa6a0d1e209131cea25cd8 |
| QACG (BERT-base) | 73.2 | Not recorded | 68fa6a0d1e209131cea25cd8 |
| LM as Fact Checker | 71.2 | Not recorded | 68fa6a0d1e209131cea25cd8 |
| QACG (RoBERTa-large) | 64.6 | Not recorded | 68fa6a0d1e209131cea25cd8 |
| LM as Fact Checker | 64.3 | Not recorded | 68fa6a0d1e209131cea25cd8 |
| MNLI-Transfer | 62.2 | Not recorded | 68fa6a0d1e209131cea25cd8 |
| QACG (BERT-base) | 56.5 | Not recorded | 68fa6a0d1e209131cea25cd8 |
| GPT2 Perplexity | 52.7 | Not recorded | 68fa6a0d1e209131cea25cd8 |
| Random Guess | 50 | Not recorded | 68fa6a0d1e209131cea25cd8 |
| MNLI-Transfer | 41.4 | Not recorded | 68fa6a0d1e209131cea25cd8 |
| GPT2 Perplexity | 35.3 | Not recorded | 68fa6a0d1e209131cea25cd8 |
| Random Guess | 33.3 | Not recorded | 68fa6a0d1e209131cea25cd8 |

### RoBERTa-large

Precision: 88. Result ID: 698e065dffc3901de09a3b20.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6a0d1e209131cea25cd8.md>)

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

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

Recorded evaluation configuration: {"Learning Setting":"Supervised","Backbone":"RoBERTa-large"}. This record does not assert independent reproduction.

### BERT-base

Precision: 85.7. Result ID: 698e065dffc3901de09a3b1f.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6a0d1e209131cea25cd8.md>)

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

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

Recorded evaluation configuration: {"Learning Setting":"Supervised","Backbone":"BERT-base"}. This record does not assert independent reproduction.

### RoBERTa-large

Precision: 85.5. Result ID: 698e065dffc3901de09a3b10.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6a0d1e209131cea25cd8.md>)

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

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

Recorded evaluation configuration: {"Learning Setting":"Supervised","Backbone":"RoBERTa-large"}. This record does not assert independent reproduction.

### BERT-base

Precision: 81.5. Result ID: 698e065dffc3901de09a3b0f.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6a0d1e209131cea25cd8.md>)

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

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

Recorded evaluation configuration: {"Learning Setting":"Supervised","Backbone":"BERT-base"}. This record does not assert independent reproduction.

### QACG (RoBERTa-large)

Precision: 77.3. Result ID: 698e065dffc3901de09a3b16.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6a0d1e209131cea25cd8.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6a0d1e209131cea25cd8.md>)

[Implementation](<https://github.com/teacherpeterpan/Zero-shot-Fact-Verification>)

Recorded evaluation configuration: {"Learning Setting":"Zero-shot","Backbone":"RoBERTa-large"}. This record does not assert independent reproduction.

### QACG (BERT-base)

Precision: 73.2. Result ID: 698e065dffc3901de09a3b15.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6a0d1e209131cea25cd8.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6a0d1e209131cea25cd8.md>)

[Implementation](<https://github.com/teacherpeterpan/Zero-shot-Fact-Verification>)

Recorded evaluation configuration: {"Learning Setting":"Zero-shot","Backbone":"BERT-base"}. This record does not assert independent reproduction.

### LM as Fact Checker

Precision: 71.2. Result ID: 698e065dffc3901de09a3b14.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6a0d1e209131cea25cd8.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa680d1e209131cea19dfa.md>)

Recorded evaluation configuration: {"Learning Setting":"Zero-shot"}. This record does not assert independent reproduction.

### QACG (RoBERTa-large)

Precision: 64.6. Result ID: 698e065dffc3901de09a3b26.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6a0d1e209131cea25cd8.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6a0d1e209131cea25cd8.md>)

[Implementation](<https://github.com/teacherpeterpan/Zero-shot-Fact-Verification>)

Recorded evaluation configuration: {"Learning Setting":"Zero-shot","Backbone":"RoBERTa-large"}. This record does not assert independent reproduction.

### LM as Fact Checker

Precision: 64.3. Result ID: 698e065dffc3901de09a3b24.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6a0d1e209131cea25cd8.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa680d1e209131cea19dfa.md>)

Recorded evaluation configuration: {"Learning Setting":"Zero-shot"}. This record does not assert independent reproduction.

### MNLI-Transfer

Precision: 62.2. Result ID: 698e065dffc3901de09a3b13.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6a0d1e209131cea25cd8.md>)

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

Recorded evaluation configuration: {"Learning Setting":"Zero-shot"}. This record does not assert independent reproduction.

### QACG (BERT-base)

Precision: 56.5. Result ID: 698e065dffc3901de09a3b25.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6a0d1e209131cea25cd8.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6a0d1e209131cea25cd8.md>)

[Implementation](<https://github.com/teacherpeterpan/Zero-shot-Fact-Verification>)

Recorded evaluation configuration: {"Learning Setting":"Zero-shot","Backbone":"BERT-base"}. This record does not assert independent reproduction.

### GPT2 Perplexity

Precision: 52.7. Result ID: 698e065dffc3901de09a3b12.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6a0d1e209131cea25cd8.md>)

Recorded evaluation configuration: {"Learning Setting":"Zero-shot"}. This record does not assert independent reproduction.

### Random Guess

Precision: 50. Result ID: 698e065dffc3901de09a3b11.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6a0d1e209131cea25cd8.md>)

Recorded evaluation configuration: {"Learning Setting":"Zero-shot"}. This record does not assert independent reproduction.

### MNLI-Transfer

Precision: 41.4. Result ID: 698e065dffc3901de09a3b23.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6a0d1e209131cea25cd8.md>)

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

Recorded evaluation configuration: {"Learning Setting":"Zero-shot"}. This record does not assert independent reproduction.

### GPT2 Perplexity

Precision: 35.3. Result ID: 698e065dffc3901de09a3b22.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6a0d1e209131cea25cd8.md>)

Recorded evaluation configuration: {"Learning Setting":"Zero-shot"}. This record does not assert independent reproduction.

### Random Guess

Precision: 33.3. Result ID: 698e065dffc3901de09a3b21.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6a0d1e209131cea25cd8.md>)

Recorded evaluation configuration: {"Learning Setting":"Zero-shot"}. This record does not assert independent reproduction.

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