Categories Companies Research

Loading the SOTA2 catalog…

Explore

Categories Companies Research

Contribute

Join SOTA2 Contact

SOTA2

© 2026 SOTA2

Build in Bellevue, USA

WizWand is now SOTA2 Research

Papers Benchmarks Datasets Tasks

Dialogue Generation research benchmarks – Page 2 · SOTA2 Research

Dialogue Generation

Benchmarks

Dataset NameSOTA MethodSort

Dataset NameSOTA methodMetricTrendResultsLast Updated

Cognitive stimulation simulated dataset 1.0 (test)

26.92ROUGE-L

12

Mar 12, 2026

22.98ROUGE-L

12

Feb 26, 2026

DuRecDial OOD (test)

4.81Coherence

11

May 13, 2026

PERSONA-CHAT Revised (dev)

85Hits@1

11

Feb 26, 2026

64.81BLEU

10

Feb 26, 2026

Commonsense Dialogue Dataset (test)

0.062Dist-1

10

Feb 26, 2026

17.4Token-level F1

9

May 20, 2026

Dialogue-level evaluation Ethical and Affective Alignment

ETHICMIND_GPT-4o

8.174Respectful Tone

9

Apr 13, 2026

Commonsense Dialogue (CD)

11.67BLEU-1

9

Feb 26, 2026

ByteDance Customer Service (test)

38.45BLEU

8

Jun 9, 2026

ByteDance Proactive Sale (test)

47.49BLEU

8

Jun 9, 2026

Human-in-the-loop Interactive Evaluation Customer-Agent Dialogs

41Win Rate (vs GPT-4)

8

Feb 26, 2026

full-hh-rlhf (test)

79.3Win Rate (Beaver-7b-v3.0-reward)

8

Feb 26, 2026

Temperature (DDS)

0.441Distinct-1

8

Feb 26, 2026

0.093BLEU-1

8

Feb 26, 2026

19.05BLEU-1

8

Feb 26, 2026

54.5MRR

8

Feb 26, 2026

SwanBench-Speech Dialogue

0.92Timbre

7

Jun 1, 2026

ZipVoice-Dia zh (test)

Ming-Flash-Omni

2.12CER (zh)

7

Mar 27, 2026

PersonalDialog Biased (test)

19.5PPL

7

Feb 26, 2026

PersonalDialog Random (test)

18.5PPL

7

Feb 26, 2026

Judicial Dialogue Human Evaluation (test)

3.99CS

6

May 15, 2026

Controlled DP (Deep)

34.7Attribute Relevancy

6

Feb 27, 2026

Commonsense Conversation Dataset

4.9BLEU

6

Feb 26, 2026

Controlled DP (Deep)

41.3Persona Controllability

6

Feb 27, 2026

Page 2 of 4

Popular datasets

DailyDialog DuRecDial Wizard of Wikipedia Persona-Chat ConvAI2

Follow for updates

Join SOTA2 Contact