Emotionally Intelligent LLM Agents

# Emotionally Intelligent LLM Agents ## Description This project aims to test the hypothesis that integrating emotional intelligence into LLM agents within the GenEnv framework enhances adaptability and performance in emotionally nuanced environments. ## Research Hypothesis Integrating emotional intelligence into LLM agents within the GenEnv framework enhances adaptability and performance in emotionally nuanced environments. ## Implementation Approach We will develop an extension of the GenEnv framework where environments simulate emotionally charged scenarios. LLM agents will be equipped with an emotional recognition module trained using sentiment analysis datasets and dynamic emotional feedback. ## Setup Instructions 1. Clone the repository: ```bash git clone cd EmotionallyIntelligentLLM ``` 2. Install the required packages: ```bash pip install -r requirements.txt ``` ## Usage Examples ### Training To train the LLM agent with emotional intelligence: ```bash python src/train.py ``` ### Evaluation To evaluate the trained agent: ```bash python src/evaluate.py ``` ## Expected Results We expect the emotionally intelligent LLM agents to show improved adaptability and decision-making in emotionally nuanced environments compared to baseline GenEnv-trained agents. ## References - GenEnv: Difficulty-Aligned Co-Evolution Between LLM Agents and Environment Simulators. [arXiv link](http://arxiv.org/abs/2512.19682v1)