From the papers that I've read on LLMs in the past 6 months, one thing is clear: higher data quality will be key to keep pushing progress.
Lots of companies and researchers keep innovating and implementing ways to improve data quality in all areas ranging from finetuning LLMs…
12
35
230
Replies
@omarsar0
Agreed, *but* people don't want to work on data quality. From a Google Research paper 2 years ago, even in incredibly high stakes use cases, most teams aren't cleaning up their data.
“Everyone wants to do the model work, not the data work” new Google research finds that under-appreciation of data quality, including in high-stakes AI, results in 92% of AI projects experiencing data cascades: compounding, negative, downstream events
5
139
471
0
4
11
@omarsar0
No doubt about it. It is a premise of more than 10 years ago, which I believe will continue to be valid for a long time.
0
0
1
@omarsar0
Yeah beyond a doubt, ‘text book is all you need’ proved that high quality data can make a big impact.
0
1
2
