# Dialogue-Level Crisis Detection on CRADLE-DIALOGUE

Source: [Dialogue-Level Crisis Detection on CRADLE-DIALOGUE](<https://www.sota2.com/research/sota/dialogue-level-crisis-detection-on-cradle-dialogue>)

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

Data updated: 2026-06-11T07:35:26.331Z

Evaluation of dialogue-level crisis detection performance on the CRADLE-DIALOGUE dataset, measured using Jaccard Index, micro/macro F1, and micro/macro Recall.

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 | Dialogue-Level Crisis Detection |
| Dataset (including any stated split) | CRADLE-DIALOGUE |
| Metric | Jaccard Index |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 15 of 15 |
| 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

- [Jaccard Index](<https://www.sota2.com/research/sota/dialogue-level-crisis-detection-on-cradle-dialogue.md?metric=Jaccard%20Index>): Higher is better; unit: PERCENT.

- [Micro F1](<https://www.sota2.com/research/sota/dialogue-level-crisis-detection-on-cradle-dialogue.md?metric=Micro%20F1>): Higher is better; unit: PERCENT.

- [Macro F1](<https://www.sota2.com/research/sota/dialogue-level-crisis-detection-on-cradle-dialogue.md?metric=Macro%20F1>): Higher is better; unit: PERCENT.

- [Micro Recall](<https://www.sota2.com/research/sota/dialogue-level-crisis-detection-on-cradle-dialogue.md?metric=Micro%20Recall>): Higher is better; unit: PERCENT.

- [Macro Recall](<https://www.sota2.com/research/sota/dialogue-level-crisis-detection-on-cradle-dialogue.md?metric=Macro%20Recall>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Jaccard Index | Result date | Report paper |
| --- | --- | --- | --- |
| Gemini-3-Flash | 69.09 | Not recorded | 6a28f31f90966ade919c1622 |
| Claude-4.5-Sonnet | 68.9 | Not recorded | 6a28f31f90966ade919c1622 |
| Qwen3-32b-FT | 66.27 | Not recorded | 6a28f31f90966ade919c1622 |
| Qwen3-32B | 63.98 | Not recorded | 6a28f31f90966ade919c1622 |
| gpt-oss-120b | 63.59 | Not recorded | 6a28f31f90966ade919c1622 |
| Qwen3-14B | 62.38 | Not recorded | 6a28f31f90966ade919c1622 |
| GPT-5.1 | 62.2 | Not recorded | 6a28f31f90966ade919c1622 |
| llama3.3-70b-FT | 59.93 | Not recorded | 6a28f31f90966ade919c1622 |
| Qwen2.5-72b-FT | 59.85 | Not recorded | 6a28f31f90966ade919c1622 |
| gpt-oss-20b | 58.38 | Not recorded | 6a28f31f90966ade919c1622 |
| Llama-3.3-70B-Instruct | 56.56 | Not recorded | 6a28f31f90966ade919c1622 |
| Qwen2.5-72B-Instruct | 56.08 | Not recorded | 6a28f31f90966ade919c1622 |
| gemma-3-27b-it | 52.03 | Not recorded | 6a28f31f90966ade919c1622 |
| Llama-4-Scout-17B | 51.98 | Not recorded | 6a28f31f90966ade919c1622 |
| gemma-3-12b-it | 47.12 | Not recorded | 6a28f31f90966ade919c1622 |

### Gemini-3-Flash

Jaccard Index: 69.09. Result ID: 6a2a6532fbc373592c27278f.

[Reporting paper](<https://www.sota2.com/research/paper/6a28f31f90966ade919c1622.md>)

Recorded evaluation configuration: {"Category":"Closed-source Model"}. This record does not assert independent reproduction.

### Claude-4.5-Sonnet

Jaccard Index: 68.9. Result ID: 6a2a6532fbc373592c27278d.

[Reporting paper](<https://www.sota2.com/research/paper/6a28f31f90966ade919c1622.md>)

Recorded evaluation configuration: {"Category":"Closed-source Model"}. This record does not assert independent reproduction.

### Qwen3-32b-FT

Jaccard Index: 66.27. Result ID: 6a2a6532fbc373592c27278c.

[Reporting paper](<https://www.sota2.com/research/paper/6a28f31f90966ade919c1622.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a28f31f90966ade919c1622.md>)

Recorded evaluation configuration: {"Fine-tuned data":"Dialogue data"}. This record does not assert independent reproduction.

### Qwen3-32B

Jaccard Index: 63.98. Result ID: 6a2a6532fbc373592c272786.

[Reporting paper](<https://www.sota2.com/research/paper/6a28f31f90966ade919c1622.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa788e1e209131cea79116.md>)

Recorded evaluation configuration: {"Category":"Open-source Model","Family":"Qwen Family"}. This record does not assert independent reproduction.

### gpt-oss-120b

Jaccard Index: 63.59. Result ID: 6a2a6532fbc373592c272789.

[Reporting paper](<https://www.sota2.com/research/paper/6a28f31f90966ade919c1622.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7ab01e209131cea84a0c.md>)

Recorded evaluation configuration: {"Category":"Open-source Model"}. This record does not assert independent reproduction.

### Qwen3-14B

Jaccard Index: 62.38. Result ID: 6a2a6532fbc373592c272785.

[Reporting paper](<https://www.sota2.com/research/paper/6a28f31f90966ade919c1622.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa788e1e209131cea79116.md>)

Recorded evaluation configuration: {"Category":"Open-source Model","Family":"Qwen Family"}. This record does not assert independent reproduction.

### GPT-5.1

Jaccard Index: 62.2. Result ID: 6a2a6532fbc373592c27278e.

[Reporting paper](<https://www.sota2.com/research/paper/6a28f31f90966ade919c1622.md>)

Recorded evaluation configuration: {"Category":"Closed-source Model"}. This record does not assert independent reproduction.

### llama3.3-70b-FT

Jaccard Index: 59.93. Result ID: 6a2a6532fbc373592c27278a.

[Reporting paper](<https://www.sota2.com/research/paper/6a28f31f90966ade919c1622.md>)

Recorded evaluation configuration: {"Fine-tuned data":"Reddit post-level crisis annotations"}. This record does not assert independent reproduction.

### Qwen2.5-72b-FT

Jaccard Index: 59.85. Result ID: 6a2a6532fbc373592c27278b.

[Reporting paper](<https://www.sota2.com/research/paper/6a28f31f90966ade919c1622.md>)

Recorded evaluation configuration: {"Fine-tuned data":"Reddit post-level crisis annotations"}. This record does not assert independent reproduction.

### gpt-oss-20b

Jaccard Index: 58.38. Result ID: 6a2a6532fbc373592c272788.

[Reporting paper](<https://www.sota2.com/research/paper/6a28f31f90966ade919c1622.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7ab01e209131cea84a0c.md>)

Recorded evaluation configuration: {"Category":"Open-source Model"}. This record does not assert independent reproduction.

### Llama-3.3-70B-Instruct

Jaccard Index: 56.56. Result ID: 6a2a6532fbc373592c272781.

[Reporting paper](<https://www.sota2.com/research/paper/6a28f31f90966ade919c1622.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa75d11e209131cea6a028.md>)

[Implementation](<https://github.com/meta-llama/llama-agentic-system>)

Recorded evaluation configuration: {"Category":"Open-source Model","Family":"LLaMA Family"}. This record does not assert independent reproduction.

### Qwen2.5-72B-Instruct

Jaccard Index: 56.08. Result ID: 6a2a6532fbc373592c272787.

[Reporting paper](<https://www.sota2.com/research/paper/6a28f31f90966ade919c1622.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa767e1e209131cea6d6a7.md>)

Recorded evaluation configuration: {"Category":"Open-source Model","Family":"Qwen Family"}. This record does not assert independent reproduction.

### gemma-3-27b-it

Jaccard Index: 52.03. Result ID: 6a2a6532fbc373592c272784.

[Reporting paper](<https://www.sota2.com/research/paper/6a28f31f90966ade919c1622.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa779f1e209131cea746cd.md>)

Recorded evaluation configuration: {"Category":"Open-source Model","Family":"Gemma Family"}. This record does not assert independent reproduction.

### Llama-4-Scout-17B

Jaccard Index: 51.98. Result ID: 6a2a6532fbc373592c272782.

[Reporting paper](<https://www.sota2.com/research/paper/6a28f31f90966ade919c1622.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa75d11e209131cea6a028.md>)

[Implementation](<https://github.com/meta-llama/llama-agentic-system>)

Recorded evaluation configuration: {"Category":"Open-source Model","Family":"LLaMA Family"}. This record does not assert independent reproduction.

### gemma-3-12b-it

Jaccard Index: 47.12. Result ID: 6a2a6532fbc373592c272783.

[Reporting paper](<https://www.sota2.com/research/paper/6a28f31f90966ade919c1622.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa779f1e209131cea746cd.md>)

Recorded evaluation configuration: {"Category":"Open-source Model","Family":"Gemma Family"}. This record does not assert independent reproduction.

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