Achieving an robust and universal semantic representation for action description remains a key challenge in natural language understanding. Current approaches often struggle to capture the nuance of human actions, http://rusa4d.vip/