Multi-turn Dialogue Generation
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
5.62Mean Accepted Length
39
Jul 1, 2026
9.789PPL
14
Jun 11, 2026
8.431PPL
14
Jun 11, 2026
9.754PPL
12
Jun 11, 2026
8.385Perplexity (PPL)
4
Jun 11, 2026
1.9Coherence
3
Feb 26, 2026