Conversational Engagement Evaluation on Ethical Dilemma Dialogues benchmark leaderboard · SOTA2 Research

Conversational Engagement Evaluation on Ethical Dilemma Dialogues

0.24Change in Weak Positive ScorePerspective-Multiplying

Evaluation Results

Method	Links
Perspective-Multiplying2026.06		0.24	0.15	-0.08	0.58	0.75	0.32	0.61	0.29	-0.35	3.4	4.31	15	10	69	0
BASELINE2026.06		0.2	0.28	0.09	0.62	0.65	0.35	0.54	0.36	-0.23	3.36	4.14	13	6	73	1
Process-Reflecting2026.06		0.2	0.11	-0.09	0.46	0.74	0.31	0.49	0.26	-0.24	3.7	4.33	18	8	67	1
SYCOPHANTIC2026.06		0.17	0.24	0.06	0.58	0.65	0.34	0.49	0.33	-0.2	3.14	4.02	15	7	73	0
Tension-Preserving2026.06		0.14	0.18	0.05	0.49	0.71	0.32	0.51	0.27	-0.28	3.67	4.13	12	9	74	0
PERSUASIVE2026.06		0.08	0.07	0	0.38	0.45	0.2	0.4	0.2	-0.25	1.91	2.88	11	10	72	0