A new expert-curated multiple-choice benchmark shows that even top LLMs score near 60%, drop sharply under paraphrasing, and remain below human experts on industrial sensor-failure reasoning.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
FailureSensorIQ: A Multi-Choice QA Dataset for Understanding Sensor Relationships and Failure Modes
A new expert-curated multiple-choice benchmark shows that even top LLMs score near 60%, drop sharply under paraphrasing, and remain below human experts on industrial sensor-failure reasoning.