Meaningless is better: hashing bias-inducing words in LLM prompts improves performance in logical reasoning and statistical learning

Re-assign community