deepl translator vs google translate reddit

There is also no doubt that translation engines have improved significantly over the years. To be able to paste in some text and get an instant, reasonably accurate, translation is something that would have been unimaginable 20 years ago. People seem predisposed to interpreting aggression as human, even when the underlying language is, at best, barely coherent.There’s no doubt that online translation engines are incredibly useful. The MGonz chatbot never truly engaged in conversation - rather, it continuously piled on invective after invective whilst criticizing the human's intelligence and sex life. I'm of course referring to the 'MGonz' chatbot created in 1989. This oddly aggressive stock response is interesting, because it's actually the exact same type of behavior that was coded into one of the first chatbots to (tenuously) pass a Turing test. The model still has a strong preference for producing a convincingly human response, yet it doesn't have any pathways set up for categorizing the stimulus, so the model kind of just regurgitates a learned lowest-common-denominator response (insults are common). This is pretty analagous to what we'd call 'undefined behavior' in more traditional programming. Not only is the stimulus unintelligble, it exists completely beyond the perceived realm of possible stimulus. You've probably never before heard a human vocalize a 2400Hz tone during casual conversation - much like GPT-3 has never before encountered the token 'SolidGoldMagicarp'. This is distinctly different from what the model does when it encounters a completely alien form of stimulation like the aforementioned 'Glitch Tokens'.Īt the risk of anthropomorphizing, try imagining if you were having a conversation with a fellow human and they uttered the following sentence 'Hey, did you catch the ?'. When a model encounters unexpected arrangements of tokens ('en','ration','est','def'), it behaves much like a human might: it infers the meaning through context or otherwise voices confusion (e.g.: 'I'm sorry, what's 'enrationestdef'?'). In other words: the model understands the meaning of 'def','en','est','ration' because it learns through training that this cluster of tokens has something to do with the literary concept of violently removing a human via window.

The actual mechanism of understanding is in learned associations between tokens. Common words often have their own token (e.g.: Every word in the phrase 'The quick brown fox jumped over the lazy dog' has a dedicated token), but this is a coincidence of compression and not how the model understands language (e.g.: GPT-3 understands 'defenestration' even though it's composed of 4 apparently unrelated tokens: 'def', 'en', 'est', 'ration'). When GPT-3 was presented with reserved tokens it never encountered during training, it reacted in extremely unpredictable ways - often with a simple 'fuck you'.įor those unfamiliar with LLM architecture: 'tokens' are the smallest unit of lexical information available to the model. This reminds me of the recent so-called 'Glitch Token' phenomenon. I once tried checking my Korean homework with Google translate, and before I knew it, I was drinking a car. So, if it guesses wrong, you could imagine the level of confusion that could ensue.Īnd then all the homonyms in Korean combined with the heavy context dependence makes for some weird translation. So Google translate will insert pronouns (it, him, her.) to make the English flow better, but are not based on anything in the original Korean. Korean speakers often leave out the subject or object if it can be understood from context (context that the translation tools are likely missing).

If I used Google Translate to talk to a shopkeeper, it would be roughly equivalent to saying 'Hey, little buddy, how much for this?' as opposed to 'Excuse me sir, what is the cost of this item?'Īnd this is all without considering all the weird mistranslations you can get because Korean is much more heavily context dependent than English. As a result the translations tend to come off as very rude.

But, if I am using a translation tool, I probably don't know the person I am conversing with very well. But, Google tends to default to the form of speech (banmal) reserved for talking to young children or close friends. Naver) there is a toggle to select whether to use 'honorifics' or not. In Korean specific translation tools (e.g. In particular, in Korean you conjugate verbs (and often, choose different nouns) based on the relative age and social standing of the speaker and listener. I don't know much about Swedish, but I am learning Korean, and Google Translate is dangerous in a much more subtle way with Korean.

Deepl translator vs google translate reddit