# Multi-user Dialogue Response Generation on PAT Source: [Multi-user Dialogue Response Generation on PAT]() Collection index: [llms.txt]() Data updated: 2026-04-29T07:21:44.561Z Evaluation of multi-user dialogue response generation performance on the PAT dataset, utilizing metrics such as Precision Coverage (P-Cov), Accuracy Coverage (A-Cov), BLEU (BL), Response Generation (RG), and Distinctness (Dist). 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 | Multi-user Dialogue Response Generation | | Dataset (including any stated split) | PAT | | Metric | P-Cov | | 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 - [P-Cov](): Higher is better; unit: PERCENT. - [A-Cov](): Higher is better; unit: PERCENT. - [BLEU-1](): Higher is better; unit: PERCENT. - [BLEU-2](): Higher is better; unit: PERCENT. - [BLEU-3](): Higher is better; unit: PERCENT. - [BLEU-4](): Higher is better; unit: PERCENT. - [RG-1](): Higher is better; unit: PERCENT. - [RG-2](): Higher is better; unit: PERCENT. - [RG-L](): Higher is better; unit: PERCENT. - [Dist-1](): Higher is better; unit: PERCENT. ## Recorded results | Method | P-Cov | Result date | Report paper | | --- | --- | --- | --- | | LLaMA-70B | 46.3 | Not recorded | 69f1907855a3c32af89fb9cf | | LLaMA-70B | 43.5 | Not recorded | 69f1907855a3c32af89fb9cf | | Anthropic Claude | 42.4 | Not recorded | 69f1907855a3c32af89fb9cf | | LLaMA-70B | 42 | Not recorded | 69f1907855a3c32af89fb9cf | | Anthropic Claude | 40.8 | Not recorded | 69f1907855a3c32af89fb9cf | | Anthropic Claude | 39.4 | Not recorded | 69f1907855a3c32af89fb9cf | | OpenAI GPT-4o | 37.4 | Not recorded | 69f1907855a3c32af89fb9cf | | Google Gemini | 37 | Not recorded | 69f1907855a3c32af89fb9cf | | OpenAI GPT-3.5 | 35.3 | Not recorded | 69f1907855a3c32af89fb9cf | | Google Gemini | 35.2 | Not recorded | 69f1907855a3c32af89fb9cf | | OpenAI GPT-3.5 | 33.3 | Not recorded | 69f1907855a3c32af89fb9cf | | Google Gemini | 33.1 | Not recorded | 69f1907855a3c32af89fb9cf | | OpenAI GPT-4o | 30.5 | Not recorded | 69f1907855a3c32af89fb9cf | | OpenAI GPT-3.5 | 30.1 | Not recorded | 69f1907855a3c32af89fb9cf | | OpenAI GPT-4o | 30.1 | Not recorded | 69f1907855a3c32af89fb9cf | ### LLaMA-70B P-Cov: 46.3. Result ID: 69f1b14a47f34ce924d7c316. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Persona":"Adaptive"}. This record does not assert independent reproduction. ### LLaMA-70B P-Cov: 43.5. Result ID: 69f1b14a47f34ce924d7c315. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Persona":"Constant"}. This record does not assert independent reproduction. ### Anthropic Claude P-Cov: 42.4. Result ID: 69f1b14a47f34ce924d7c30a. [Reporting paper]() Recorded evaluation configuration: {"Persona":"Adaptive"}. This record does not assert independent reproduction. ### LLaMA-70B P-Cov: 42. Result ID: 69f1b14a47f34ce924d7c314. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Persona":"w/o Persona"}. This record does not assert independent reproduction. ### Anthropic Claude P-Cov: 40.8. Result ID: 69f1b14a47f34ce924d7c309. [Reporting paper]() Recorded evaluation configuration: {"Persona":"Constant"}. This record does not assert independent reproduction. ### Anthropic Claude P-Cov: 39.4. Result ID: 69f1b14a47f34ce924d7c308. [Reporting paper]() Recorded evaluation configuration: {"Persona":"w/o Persona"}. This record does not assert independent reproduction. ### OpenAI GPT-4o P-Cov: 37.4. Result ID: 69f1b14a47f34ce924d7c313. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Persona":"Adaptive"}. This record does not assert independent reproduction. ### Google Gemini P-Cov: 37. Result ID: 69f1b14a47f34ce924d7c30d. [Reporting paper]() Recorded evaluation configuration: {"Persona":"Adaptive"}. This record does not assert independent reproduction. ### OpenAI GPT-3.5 P-Cov: 35.3. Result ID: 69f1b14a47f34ce924d7c310. [Reporting paper]() Recorded evaluation configuration: {"Persona":"Adaptive"}. This record does not assert independent reproduction. ### Google Gemini P-Cov: 35.2. Result ID: 69f1b14a47f34ce924d7c30b. [Reporting paper]() Recorded evaluation configuration: {"Persona":"w/o Persona"}. This record does not assert independent reproduction. ### OpenAI GPT-3.5 P-Cov: 33.3. Result ID: 69f1b14a47f34ce924d7c30e. [Reporting paper]() Recorded evaluation configuration: {"Persona":"w/o Persona"}. This record does not assert independent reproduction. ### Google Gemini P-Cov: 33.1. Result ID: 69f1b14a47f34ce924d7c30c. [Reporting paper]() Recorded evaluation configuration: {"Persona":"Constant"}. This record does not assert independent reproduction. ### OpenAI GPT-4o P-Cov: 30.5. Result ID: 69f1b14a47f34ce924d7c312. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Persona":"Constant"}. This record does not assert independent reproduction. ### OpenAI GPT-3.5 P-Cov: 30.1. Result ID: 69f1b14a47f34ce924d7c30f. [Reporting paper]() Recorded evaluation configuration: {"Persona":"Constant"}. This record does not assert independent reproduction. ### OpenAI GPT-4o P-Cov: 30.1. Result ID: 69f1b14a47f34ce924d7c311. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Persona":"w/o Persona"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.