Curiosity-driven dialogue evaluation on Claude-based dialogues benchmark leaderboard · SOTA2 Research

Curiosity-driven dialogue evaluation on Claude-based dialogues

3.29L1 ScoreCURIOBOT

Evaluation Results

Method	Links
CURIOBOT2026.06		3.29	3.02	3.63	3.54	5.33	4.19	4.58	4.76	5.27	4.35
Baseline2026.06		2.72	2.06	2.88	2.73	5.28	3.07	3.88	3.82	3.85	2.94