Am I missing something? Most of TFA is about Bard failing to answer with rhyming words, but in the only prompts shown the author doesn't actually ask for rhyming words. He just says the hint and the name of the puzzle.
Is this not simply: "Bard is worse than ChatGPT at having seen the 'how-to-play' page for my side project during its training"?
How is this possible? Google makes people do 8 rounds of leetcode. How could they be beaten? Nothing makes sense anymore.
It's so sad to me to see the downfall of google from the absolute coolest company on the planet to the one that's now trying to keep up.
That's a clever game to get it to play. Today I asked ChatGPT to give me 1000 Fibonacci numbers starting with the 2000th number and it crashed. Later I asked it the same prompt and it repeatedly gave me code to calculate the Fibonacci numbers in Python.
> Twofer Goofer HQ's adherence to strict "perfect" rhyme can be tricky for those slant rhyme-inclined.
And yet one puzzle they hammer Bard for failing is "Cactus Practice". What accent do you have to have for that to be a perfect rhyme?
Offtopic - have you seen Phind?
It is very fast and wins the search benchmarks here:
If I understand how Large Language models work, they don't actually know about spelling.... they are given tokens that represent words, and can only infer things from the context of those tokens across terabytes of data that they're given.
Any rhyming done is an impressive result.
"Bard is much worse than ChatGPT at solving an obscure word game I invented" would have been a more honest title, but would probably generate less clicks for the author.
Bard may still be much worse than ChatGPT at solving all kinds of puzzles, but the article is click bait for promoting the author's word game, not an actual investigation that warrants that conclusion.
How do you navigate this blog to read the other articles? I couldn't find any way to read the one on gpt4 (clicking the underlined "wrote about" does nothing) and twofergoofer.com/blog goes to a 404.
It is interesting how lackluster the reactions are about Bard, when it would have been jaw-gapping amazing just a year ago.
Bard is basically trash
I tried to play hangman with it, but it was on crack.
[dead]
Wow I had hoped for a more productive discussion than these 1-1 comparisons of Bard vs ChatGPT that I'm seeing everywhere. The model deployed with this version of Bard is clearly a smaller model than the biggest LaMDA/PaLM models Google has been working on for ages. Which, according to their publications, show unprecedented results on _proof writing_ of all things (see Minerva). While their strategic decisions may be questionable (or they're just trying to quantize the model for mass deployment without burning billions per month in compute costs), its almost silly to question Google's ability to build useful LLMs.