# Dialogue Generation on Cognitive stimulation real 1.0 (test) Source: [Dialogue Generation on Cognitive stimulation real 1.0 (test)]() Collection index: [llms.txt]() Data updated: 2026-03-12T07:13:29.409Z Evaluation of Dialogue Generation performance on the test split of the Cognitive stimulation real 1.0 dataset, utilizing metrics covering overlap (ROUGE, BLEU), semantic similarity (BERTScore), diversity (Distinct-2), and subjective quality (Relevance, Empathy, Fluency). 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 Generation | | Dataset (including any stated split) | Cognitive stimulation real 1.0 (test) | | Metric | ROUGE-L | | Unit | PERCENT | | Direction | Higher 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 - [ROUGE-L](): Higher is better; unit: PERCENT. - [BLEU-2](): Higher is better; unit: PERCENT. - [BLEU-4](): Higher is better; unit: PERCENT. - [BERTScore](): Higher is better; unit: PERCENT. - [Distinct-2](): Higher is better; unit: PERCENT. - [Relevance](): Higher is better; unit: SCALAR. - [Empathy](): Higher is better; unit: SCALAR. - [Fluency](): Higher is better; unit: SCALAR. ## Recorded results | Method | ROUGE-L | Result date | Report paper | | --- | --- | --- | --- | | GCSD-3b | 27.63 | Not recorded | 69b24864d030084c22e2f71f | | GPT-4o | 25.76 | Not recorded | 69b24864d030084c22e2f71f | | GCSD | 25.18 | Not recorded | 69b24864d030084c22e2f71f | | Doubao-Pro | 24.87 | Not recorded | 69b24864d030084c22e2f71f | | GCSD | 24.33 | Not recorded | 69b24864d030084c22e2f71f | | LLaMA-3.1-405B | 23.87 | Not recorded | 69b24864d030084c22e2f71f | | DeepSeek-671B | 23.58 | Not recorded | 69b24864d030084c22e2f71f | | GCSD | 23.57 | Not recorded | 69b24864d030084c22e2f71f | | Qwen-Max | 22.87 | Not recorded | 69b24864d030084c22e2f71f | | DeepSeek-671B | 22.54 | Not recorded | 69b24864d030084c22e2f71f | | ChatGPT | 22.32 | Not recorded | 69b24864d030084c22e2f71f | | GLM-4-Plus | 18.54 | Not recorded | 69b24864d030084c22e2f71f | | ERNIE | 18.32 | Not recorded | 69b24864d030084c22e2f71f | ### GCSD-3b ROUGE-L: 27.63. Result ID: 69b265df6dbb7075b95c82b7. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### GPT-4o ROUGE-L: 25.76. Result ID: 69b265df6dbb7075b95c82ad. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### GCSD ROUGE-L: 25.18. Result ID: 69b265df6dbb7075b95c82b6. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Condition":"w/o CSFAL"}. This record does not assert independent reproduction. ### Doubao-Pro ROUGE-L: 24.87. Result ID: 69b265df6dbb7075b95c82ac. [Reporting paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### GCSD ROUGE-L: 24.33. Result ID: 69b265df6dbb7075b95c82b4. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Condition":"w/o CT"}. This record does not assert independent reproduction. ### LLaMA-3.1-405B ROUGE-L: 23.87. Result ID: 69b265df6dbb7075b95c82af. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### DeepSeek-671B ROUGE-L: 23.58. Result ID: 69b265df6dbb7075b95c82b2. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Prompting":"5-Shot"}. This record does not assert independent reproduction. ### GCSD ROUGE-L: 23.57. Result ID: 69b265df6dbb7075b95c82b5. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Condition":"w/o DPSM"}. This record does not assert independent reproduction. ### Qwen-Max ROUGE-L: 22.87. Result ID: 69b265df6dbb7075b95c82b3. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Prompting":"5-Shot"}. This record does not assert independent reproduction. ### DeepSeek-671B ROUGE-L: 22.54. Result ID: 69b265df6dbb7075b95c82ae. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Prompting":"Zero-shot"}. This record does not assert independent reproduction. ### ChatGPT ROUGE-L: 22.32. Result ID: 69b265df6dbb7075b95c82b1. [Reporting paper]() Recorded evaluation configuration: {"Prompting":"5-Shot"}. This record does not assert independent reproduction. ### GLM-4-Plus ROUGE-L: 18.54. Result ID: 69b265df6dbb7075b95c82b0. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Prompting":"5-Shot"}. This record does not assert independent reproduction. ### ERNIE ROUGE-L: 18.32. Result ID: 69b265df6dbb7075b95c82ab. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.