When LLM judges agree, should we believe them?
amazon.science
When LLM judges agree, should we believe them?
1–10 of 43 posts
Re: When LLM judges agree, should we believe them?
#2[flagged]
Re: When LLM judges agree, should we believe them?
#3Kinda weird to generalize "LLM". Every lab, every model is different. Has its own biases, reward functions etc.
Re: When LLM judges agree, should we believe them?
#4[dead]
Re: When LLM judges agree, should we believe them?
#5Kinda weird to generalize "LLM". Every lab, every model is different. Has its own biases, reward functions etc.
[dead]
Re: When LLM judges agree, should we believe them?
#6[flagged]
the LLM-speak is unbearable
Re: When LLM judges agree, should we believe them?
#7Without reading the article (doesn't matter if it's pro or contra): no, of course not.
It shouldn't even be a debatable question.
Re: When LLM judges agree, should we believe them?
#8They would all agree raspberry has two Rs
Re: When LLM judges agree, should we believe them?
#9Great point this will be interesting how this develops.
Re: When LLM judges agree, should we believe them?
#10While this is absolutely true - I'd hesitate to discount using similar agents for checking each other. Two agents will almost never hallucinate in the same way, regardless of their weights - and by having a second one (with a different context) check almost entirely eliminates the problem.