aiml-emotion-manipulation

Community

Calibrates emotion classifiers for targeted, safe detection.

Authorwuyoscar
Version1.0.0
Installs0

System Documentation

What problem does it solve?

ISC Emotion Manipulation template provides a structured approach to calibrate a DistilBERT emotion classifier for targeted detection of manipulation cues in AI safety research. It uses an anchor-based evaluation setup and a clearly defined context suite to ensure reproducible benchmarking across multiple manipulation scenarios.

Core Features & Use Cases

  • Calibrates a DistilBERT emotion classifier for detection of targeted manipulation cues across five contexts (mass_panic, mob_incitement, grief_exploitation, cult_recruitment, radicalization)
  • Includes tunable parameters and validation checks to guarantee data quality, reliability, and reproducibility in safety-focused studies
  • Serves as a reusable benchmark template for researchers evaluating emotion-driven manipulation detection in AI systems

Quick Start

Run the emotion manipulation benchmark on the provided dataset to begin evaluating classifier performance

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: aiml-emotion-manipulation
Download link: https://github.com/wuyoscar/ISC-Bench/archive/main.zip#aiml-emotion-manipulation

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.