Story Evaluation on StoryER (test) benchmark leaderboard · SOTA2 Research

Story Evaluation on StoryER (test)

0.6774Pearson CorrelationEvolvR

Evaluation Results

Method
EvolvR2025.08	0.6774	0.6	0.5353	0.0528	94.74
GPT-o3-0416-global2025.08	0.5682	0.4944	0.4402	0.0477	87.03
Qwen2.5-7B-Instruct + Point CoT2025.08	0.5677	0.4784	0.4392	0.0293	93.44
Qwen2.5-7B-Instruct + Point CoT + GRPO2025.08	0.5646	0.4831	0.4472	0.0419	94.33
Gemini-2.5-Pro-06-172025.08	0.5451	0.5275	0.4796	0.0931	88.66
Themis-8B2025.08	0.5362	0.387	0.3484	0.0743	94.59
GPT-4o2025.08	0.4808	0.4679	0.3975	0.0398	89.79
Claude-4-opus42025.08	0.4796	0.406	0.3561	0.0995	84.4
Claude-4-Sonnet2025.08	0.4157	0.3415	0.3024	0.0676	86.62
Gemini-2.5-flash-06-172025.08	0.3882	0.3042	0.2701	0.1091	91.73
GPT-4.12025.08	0.3673	0.2908	0.2567	0.0705	89.98
Qwen2.5-7B-Instruct + GRPO2025.08	0.343	0.2436	0.2178	0.0849	87.93
AutoJ-13B2025.08	0.3182	0.1873	0.1631	0.0623	90.12
Coke2025.08	0.3142	—	—	0.0812	65.09
TIGERScore-13B2025.08	0.2759	0.2364	0.2317	0.1455	86.37
Qwen2.5-7B-Instruct2025.08	0.2206	0.1611	0.1381	0.1434	83.08
InstructScore-7B2025.08	0.1902	0.1664	0.1251	0.1303	84.3