Malware-as-image benchmarks leak near-duplicate samples into tests
Content-aware deduplication cuts top-1 accuracy by up to 8 points and macro-F1 by up to 15 on leaky benchmarks.
Stories per day · 30 days
1 stories, newest first · one card per story
1 stories, newest first · one card per story
Content-aware deduplication cuts top-1 accuracy by up to 8 points and macro-F1 by up to 15 on leaky benchmarks.
Today 1.6K readers · 2.4K views · 774 reads · all time 189K views