# Conversational Question Answering on CANARD

Source: [Conversational Question Answering on CANARD](<https://www.sota2.com/research/sota/conversational-question-answering-on-canard>)

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

Data updated: 2026-02-26T05:43:07.036Z

Evaluation of model performance on the CANARD dataset for Conversational Question Answering, reporting metrics such as F1, HEQ-Q, and HEQ-D.

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 | Conversational Question Answering |
| Dataset (including any stated split) | CANARD |
| Metric | F1 Score |
| Unit | SCALAR |
| Direction | Higher is better |
| Results loaded | 9 of 9 |
| 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

- [F1 Score](<https://www.sota2.com/research/sota/conversational-question-answering-on-canard.md?metric=F1%20Score>): Higher is better; unit: SCALAR.

- [HEQ-Q Score](<https://www.sota2.com/research/sota/conversational-question-answering-on-canard.md?metric=HEQ-Q%20Score>): Higher is better; unit: SCALAR.

- [HEQ-D Deviation](<https://www.sota2.com/research/sota/conversational-question-answering-on-canard.md?metric=HEQ-D%20Deviation>): Lower is better; unit: SCALAR.

## Recorded results

| Method | F1 Score | Result date | Report paper |
| --- | --- | --- | --- |
| EXCORD | 0.681 | Not recorded | 68fa6a3f1e209131cea26ff4 |
| RoBERTa | 0.669 | Not recorded | 68fa6a3f1e209131cea26ff4 |
| RoBERTa | 0.658 | Not recorded | 68fa6a3f1e209131cea26ff4 |
| EXCORD | 0.631 | Not recorded | 68fa6a3f1e209131cea26ff4 |
| EXCORD | 0.626 | Not recorded | 68fa6a3f1e209131cea26ff4 |
| BERT+HAE | 0.624 | Not recorded | 68fa6a3f1e209131cea26ff4 |
| BERT | 0.622 | Not recorded | 68fa6a3f1e209131cea26ff4 |
| BERT+HAE | 0.582 | Not recorded | 68fa6a3f1e209131cea26ff4 |
| BERT | 0.574 | Not recorded | 68fa6a3f1e209131cea26ff4 |

### EXCORD

F1 Score: 0.681. Result ID: 698e23e4ffc3901de09b1c09.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6a3f1e209131cea26ff4.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6a3f1e209131cea26ff4.md>)

[Implementation](<https://github.com/dmis-lab/excord>)

Recorded evaluation configuration: {"Backbone":"RoBERTa"}. This record does not assert independent reproduction.

### RoBERTa

F1 Score: 0.669. Result ID: 698e23e4ffc3901de09b1c08.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6a3f1e209131cea26ff4.md>)

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

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

Recorded evaluation configuration: {"Approach":"Pipeline"}. This record does not assert independent reproduction.

### RoBERTa

F1 Score: 0.658. Result ID: 698e23e4ffc3901de09b1c07.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6a3f1e209131cea26ff4.md>)

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

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

Recorded evaluation configuration: {"Approach":"End-to-end"}. This record does not assert independent reproduction.

### EXCORD

F1 Score: 0.631. Result ID: 698e23e4ffc3901de09b1c06.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6a3f1e209131cea26ff4.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6a3f1e209131cea26ff4.md>)

[Implementation](<https://github.com/dmis-lab/excord>)

Recorded evaluation configuration: {"Backbone":"BERT+HAE"}. This record does not assert independent reproduction.

### EXCORD

F1 Score: 0.626. Result ID: 698e23e4ffc3901de09b1c03.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6a3f1e209131cea26ff4.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6a3f1e209131cea26ff4.md>)

[Implementation](<https://github.com/dmis-lab/excord>)

Recorded evaluation configuration: {"Backbone":"BERT"}. This record does not assert independent reproduction.

### BERT+HAE

F1 Score: 0.624. Result ID: 698e23e4ffc3901de09b1c05.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6a3f1e209131cea26ff4.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa66901e209131cea10dea.md>)

Recorded evaluation configuration: {"Approach":"Pipeline"}. This record does not assert independent reproduction.

### BERT

F1 Score: 0.622. Result ID: 698e23e4ffc3901de09b1c02.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6a3f1e209131cea26ff4.md>)

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

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

Recorded evaluation configuration: {"Approach":"Pipeline"}. This record does not assert independent reproduction.

### BERT+HAE

F1 Score: 0.582. Result ID: 698e23e4ffc3901de09b1c04.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6a3f1e209131cea26ff4.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa66901e209131cea10dea.md>)

Recorded evaluation configuration: {"Approach":"End-to-end"}. This record does not assert independent reproduction.

### BERT

F1 Score: 0.574. Result ID: 698e23e4ffc3901de09b1c01.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6a3f1e209131cea26ff4.md>)

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

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

Recorded evaluation configuration: {"Approach":"End-to-end"}. This record does not assert independent reproduction.

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