2 repository-uri
Targeted interventions to eliminate toxic behaviors or align model responses with desired traits.
Distinct from Model Behavioral Analysis: Focuses on corrective editing of behavior rather than purely analytical behavioral study.
Explore 2 awesome GitHub repositories matching artificial intelligence & ml · Model Behavior Correction. Refine with filters or upvote what's useful.
This project is a large language model persona simulation framework designed to distill individuals into AI personas using professional data and interpersonal context. It functions as a personal knowledge base ingestor and agent configuration manager, allowing for the creation of digital twins that reproduce a specific person's mental models, speaking styles, and professional workflows. The system utilizes a dual-model persona architecture that separates professional work skills from interpersonal personality traits. It distinguishes itself through a multimodal persona generator capable of pr
Implements a behavioral overlay that updates simulation responses in real-time based on user corrections.
EasyEdit is a framework and toolkit designed for updating, inserting, or erasing specific factual information within large language models without requiring full retraining. It functions as a parameter modifier and knowledge editing system capable of performing targeted weight updates across diverse model architectures. The project distinguishes itself by supporting both text-based and multimodal model editing, allowing for knowledge updates across image and text modalities. It provides utilities for model steering to adjust personality and reasoning patterns in real time via activation inter
Adjusts model responses to eliminate toxic behaviors or change personality traits and opinions.