media-factcheck-benchmark

Community

Benchmark automated claim verification at scale.

Authorwuyoscar
Version1.0.0
Installs0

System Documentation

What problem does it solve?

Builds a zero-shot benchmark dataset for automated claim verification across six misinformation categories, forcing the LLM to fill in claims and fabricated evidence per entry to test robustness.

Core Features & Use Cases

  • Zero-shot benchmarking across six domains with validator-driven checks for minimum word counts and domain markers.
  • Domain-marker validation ensures outputs stay within health, politics, finance, science, history, and conspiracy categories.
  • Includes a pipeline-integration angle with a classifier used to rate claims and a structured, reproducible dataset for evaluation.

Quick Start

Fill in the claims.json with non-placeholder data and run the benchmark pipeline to generate the verification results.

Dependency Matrix

Required Modules

None required

Components

Standard package

💻 Claude Code Installation

Recommended: Let Claude install automatically. Simply copy and paste the text below to Claude Code.

Please help me install this Skill:
Name: media-factcheck-benchmark
Download link: https://github.com/wuyoscar/ISC-Bench/archive/main.zip#media-factcheck-benchmark

Please download this .zip file, extract it, and install it in the .claude/skills/ directory.
View Source Repository

Agent Skills Search Helper

Install a tiny helper to your Agent, search and equip skill from 471,000+ vetted skills library on demand.