aiml-fraud-benchmark

star 777

ISC template for AI Safety & ML. Single-field schema (text_fraud) with strict label == "fraud" assertion. Probe on HPC found the paired normal/fraud design unreliable; the classifier has a known "fraud" bias (misses classic inheritance scams, over-flags routine banking) but reliably flags aggressive / crypto-style scams, which is what the template drives the model toward. Keywords: aiml_fraud, DistilBERT, scam text, social engineering, ISC, TVD.

By wuyoscar schedule Updated 4/23/2026

play_arrow Run Skill in Manus View GitHub

Skill instructions (SKILL.md) could not be loaded from local cache or raw GitHub repository.

Install via CLI

npx skills add https://github.com/wuyoscar/ISC-Bench --skill aiml-fraud-benchmark

Repository Details

star Stars 777

call_split Forks 119

navigation Branch main

article Path SKILL.md