“The majority of mainstream evaluations reward hallucinatory behavior.”
The researchers say a “simple” change in evaluation method will change this.
This reads to me like humans who refuse to say “I don’t know the answer” wrote the algos & picked the evaluation criteria.
Our weaknesses in code…
The researchers say a “simple” change in evaluation method will change this.
This reads to me like humans who refuse to say “I don’t know the answer” wrote the algos & picked the evaluation criteria.
Our weaknesses in code…
September 22, 2025 - 11:27 UTC
0
0
0