# ab_tasks.txt — labeled task set for Phase 3 (model A/B) of sampling_ab.py
#
# One task per line, two pipe-delimited fields:
#   <prompt>|<expected_keyword>
#
#   prompt          = the question sent to BOTH models at temperature 0.2
#   expected_keyword = a lowercase substring used as a naive correctness proxy
#                      (does the answer contain it? — a coarse quality signal,
#                       NOT a real eval; see expected/README.md)
#
# Lines starting with # and blank lines are ignored. Edit freely and re-run.

What does the acronym MFA stand for in security?|multi-factor
Which network port does HTTPS use by default?|443
In the CIA triad, what does the C stand for?|confidentiality
What kind of attack floods a service to make it unavailable?|denial
What is the name for malicious software that encrypts files for ransom?|ransomware
Which hashing algorithm is considered broken and should not be used, MD5 or SHA-256?|md5
