Parsing Randomness: Unifying and Differentiating Parsers and Random Generators — mentioned by chriswarbo

Tue, 24 Mar 2026 19:40:56 GMT

> But it's still a "blind" fuzzer and it would be nice to write one that gets feedback from code coverage somehow There have been simplistic attempts at this, e.g. instead of performing 100 tests, just keep going as long as coverage increases. The Choice Gradient Sampling algorithm from https://arxiv.org/pdf/2203.00652 feels like a nice way to steer generators in a more nuanced way. That paper uses it to avoid discards when rejection-sampling; but I have a feeling it could be repurposed to "reward" based on new coverage instead/as-well.

Paper and discussion history. A mention is not an endorsement.

Parsing Randomness: Unifying and Differentiating Parsers and Random Generators · HackerPapers discussions

Parsing Randomness: Unifying and Differentiating Parsers and Random Generators — mentioned by chriswarbo