Positional preferences, order effects, prompt sensitivity undermine AI judgments
1–10 of 87 posts
Re: Positional preferences, order effects, prompt sensitivity undermine AI judgments
#2[flagged]
Re: Positional preferences, order effects, prompt sensitivity undermine AI judgments
#3I listen to online debates, especially political ones on various platforms, and man. The AI slop that people slap around at each other is beyond horrendous. I would not want an LLM being the final say on something critical. I want the opposite, an LLM should identify things that need follow up review by a qualified person, a person should still confirm the things that "pass" but they can then prioritize what to validate first.
Re: Positional preferences, order effects, prompt sensitivity undermine AI judgments
#4[flagged]
At least until the LLM judges otherwise.
Re: Positional preferences, order effects, prompt sensitivity undermine AI judgments
#5I’d argue real judges are unreliable as well.
The real question for me is: are they less reliable than human judges? Probably yes. But I favor a relative measurement to humans than a plain statement like that.
Re: Positional preferences, order effects, prompt sensitivity undermine AI judgments
#6Some other known distributional biases include self-preference bias (gpt-4o prefers gpt-4o generations over claude generations for eg) and structured output/JSON-mode bias [1]. Interestingly, some models have a more positive/negative-skew than others as well. This library [2] also provides some methods for calibrating/stabilizing them.
[1]: https://verdict.haizelabs.com/docs/cookbook/distributional-b... [2]: https://github.com/haizelabs/verdict
Re: Positional preferences, order effects, prompt sensitivity undermine AI judgments
#7I’d argue real judges are unreliable as well. The real question for me is: are they less reliable than human judges? Probably yes. But I favor a relative measurement to humans than a plain statement like that.
[dead]
Re: Positional preferences, order effects, prompt sensitivity undermine AI judgments
#8I listen to online debates, especially political ones on various platforms, and man. The AI slop that people slap around at each other is beyond horrendous. I would not want an LLM being the final say on something critical. I want the opposite, an LLM should identify things that need follow up review by a qualified person, a person should still confirm the things that "pass" but they can then prioritize what to valid…
I don't even trust LLMs enough to spot content that requires validation or nuance.
Re: Positional preferences, order effects, prompt sensitivity undermine AI judgments
#9I see "panels of judges" mentioned once, but what is the weakness of this? Other than more resource.
Worst case you end up with some multi-modal distribution, where two opinions are equal - which seems somewhat unlikely as the panel size grows. Or it could maybe happen in some case with exactly two outcomes (yes/no), but I'd be surprised if such a panel landed on a perfect uniform distribution in its judgments/opinions (50% yes 50% no)
Re: Positional preferences, order effects, prompt sensitivity undermine AI judgments
#10I’d argue real judges are unreliable as well. The real question for me is: are they less reliable than human judges? Probably yes. But I favor a relative measurement to humans than a plain statement like that.
There are technical quirks that make LLM judges particularly high variance, sensitive to artifacts in the prompt, and positively/negatively-skewed, as opposed to the subjectivity of human judges. These largely arise from their training distribution and post-training, and can be contained with careful calibration.