Gork@sopuli.xyz to No Stupid Questions@lemmy.world · 14 hours agoDo LLM modelers maintain a list of manual corrections fed by humans?message-squaremessage-square12fedilinkarrow-up132arrow-down12file-text
arrow-up130arrow-down1message-squareDo LLM modelers maintain a list of manual corrections fed by humans?Gork@sopuli.xyz to No Stupid Questions@lemmy.world · 14 hours agomessage-square12fedilinkfile-text
Like the how many r’s in strawberry. It took off as an Internet meme and was fixed, but how did that fix happen?
minus-squareblack_flag@lemmy.dbzer0.comlinkfedilinkarrow-up3·10 hours agoThat’s not always true, they can also use regular logic to flag certain requests (like r’s in strawberry) and respond manually without it ever reaching the model
minus-squarethreeonefour@piefed.calinkfedilinkEnglisharrow-up1·9 hours agoRight, but that’s like saying you can fix a broken space heater by wearing a sweater.
minus-squareblack_flag@lemmy.dbzer0.comlinkfedilinkarrow-up1·8 hours agoMaybe I misinterpreted the question, I was thinking this was including presentation layer for llm tools
That’s not always true, they can also use regular logic to flag certain requests (like r’s in strawberry) and respond manually without it ever reaching the model
Right, but that’s like saying you can fix a broken space heater by wearing a sweater.
Maybe I misinterpreted the question, I was thinking this was including presentation layer for llm tools