TouchStone
TouchStone copied to clipboard
Touchstone: Evaluating Vision-Language Models by Language Models
I notice that VisualGLM and Qwen-VL are evaluated on the Chinese split of TouchStone. Is the Chinese split translated from the English split? If not, could you please release the...
When I run it in the console: ```shell python3 eval.py ./touchstone_20230831.tsv sk-*** --model-name gpt-4 ``` console output: ``` -------------- evaluate gpt-4 ------------- Traceback (most recent call last): File "eval.py", line...
