../

HARM: A Hint-Based Assertion Miner


Year
2022
Authors
Samuele Germiniani, Graziano Pravadelli
DOI
10.1109/TCAD.2022.3197525

Related

Persistent Notes

In-text annotations

“While static mining pursues generalizations and abstractions, dynamic approaches can only extract likely assertions, i.e., formulas that represent solely the behaviors that are exposed in the considered execution traces. On the other hand” Page 4277

“static methods have scalability issues while dynamic assertion mining is much more computationally efficient." Page 4277

“Thus, the majority of existing dynamic assertion miners exploit heuristic approaches to guarantee scalability at the cost of a poor set of generated assertions, while other tools sacrifice scalability to provide a more complete set of assertions." Page 4277

“A very fast miner engine (with respect to the stateof-the-art). HARM features a parallelized linear-time algorithm to evaluate an assertion and generate its contingency table, efficiently managing input traces that are millions of time units long." Page 4278

%% Import Date: 2026-06-13T15:09:47.945-04:00 %%