OpenCompass/opencompass/configs/datasets/cmo_fib
Songyang Zhang f97c4eae42
[Update] Update Fullbench (#1712)
* Update JuderBench

* Support O1-style Prompts

* Update Code
2024-11-26 14:26:55 +08:00
..
cmo_fib_0shot_notcot_gen_4c6c29.py [Update] Update Fullbench (#1712) 2024-11-26 14:26:55 +08:00
cmo_fib_gen_ace24b.py [Datasets] Add datasets CMO&AIME (#1610) 2024-10-28 18:08:02 +08:00
cmo_fib_gen.py [Datasets] Add datasets CMO&AIME (#1610) 2024-10-28 18:08:02 +08:00
README.md [Datasets] Add datasets CMO&AIME (#1610) 2024-10-28 18:08:02 +08:00

Description

Math dataset composed of problems from CMO (Chinese Mathematical Olympiad) 2009-2022 .

Performance

Qwen2.5-Math-72B-Instruct Qwen2.5-Math-7B-Instruct Qwen2-Math-7B-Instruct Qwen2-Math-1.5B-Instruct internlm2-math-7b
46.15 42.79 31.73 23.56 3.37
Qwen2.5-72B-Instruct Qwen2.5-7B-Instruct internlm2_5-7b-chat
20.00 16.67 6.67