FIXME is an open 180-task benchmark for LLM functional verification, finding that models are strongest at specification questions and weakest at testcase design and assertion generation.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.AR 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
FIXME: Towards End-to-End Benchmarking of LLM-Aided Design Verification
FIXME is an open 180-task benchmark for LLM functional verification, finding that models are strongest at specification questions and weakest at testcase design and assertion generation.