CHALLENGE DESCRIPTION
Words carry semantic information. Similar to how people can infer meaning based on a word’s context, AI can derive representations for words based on their context too! However, the kinds of meaning that a model uses may not match ours. We’ve found a pair of AIs speaking in metaphors that we can’t make any sense of! The embedding model is glove-twitter-25. Note that the flag should be fully ASCII ans starts with ‘htb{’.