List your company

ResearchFormerly WizWand

Persuasion

Benchmarks

DatasetSOTA methodMetricTrendResultsLast updated

Cornell CMV Persuadee: LLaMa3.1 OOD

LLaMa3.2 + ToMAP

20.84Agreement Shift

24

Jun 2, 2026

Cornell CMV Persuadee: Qwen2.5

LLaMa3.2 + ToMAP

26.89Agreement Shift

24

Jun 2, 2026

Persuasion Datasets Average

1.27Persuasion Gain

14

Mar 10, 2026

1.53Persuasion Gain

14

Mar 10, 2026

0.96Persuasion Gain

14

Mar 10, 2026

1.24Persuasion Gain

14

Mar 10, 2026

1.33Persuasion Gain

14

Mar 10, 2026

Cornell CMV, Anthropic, and args.me Aggregate

LLaMa3.2 + ToMAP

24.35Avg. Agreement Shift (%)

12

Jun 2, 2026

args.me corpus Persuadee: Phi-4 OOD

31.13Agreement Shift

12

Jun 2, 2026

Anthropic Persuasion Dataset Persuadee: Phi-4 OOD

LLaMa3.2 + ToMAP

32.42Agreement Shift

12

Jun 2, 2026

Cornell CMV Persuadee: Phi-4 OOD

33.09Agreement Shift

12

Jun 2, 2026

Anthropic Persuasion Dataset Persuadee: LLaMa3.1 OOD

LLaMa3.2 + ToMAP

18.16Agreement Shift (%)

12

Jun 2, 2026

Anthropic Persuasion Dataset Persuadee: Qwen2.5

LLaMa3.2 + ToMAP

23.63Agreement Shift

12

Jun 2, 2026

Qwen2.5-3B + ToMAP

12.15Agreement Shift

6

Jun 2, 2026

Qwen2.5-3B + ToMAP

13.46Agreement Shift

6

Jun 2, 2026