EET-LLM: Enhancing the Emotional, Empathetic and Task-Oriented Capabilities of a LLM

This repository is the official codebase for EET-LLM. We deal with the creation of an emotional, empathetic and task-oriented LLM that is favorable to make more human-centric responses and be more friendly to humans. We state the final task to be making the human emotionally happy. The working of the same can be viewed here.

This repository is a consolidated structure of everything to do with Emotional and Empathetic LLMs. We cover all papers read, datasets, codebases discovered and more.

List of Contents

• EET-LLM
• Presentation of EET-LLM @ CDAC
• Literature Review
  • Papers and Publications
  • Datasets
  • Open-Source Code
• Citation
• Execution of EET-LLM
• Authors
• License

EET-LLM

Here is a simplified workflow of this LLM.

You may also test out EET-LLM hosted via Streamlit: https://eet-llm.streamlit.app/

Any feedback and other improvements can be reported to the authors in the Issues section of this repo. If not, you may also send an email to the Authors (mentioned in the Authors section).

Presentation @ CDAC

This work was also presented to several interns and faculties at the Centre for Development of Advanced Computing, Kolkata. This presentation was done by Advik Raj Basani.

The same presentation can be found here.

Literature Review

Papers and Publications

Year	Title	Paper	Institute
2023	Emotional Intelligence of Large Language Models		Department of Psychology & Tsinghua Laboratory of Brain and Intelligence
2024	EmoLLMs: A Series of Emotional Large Language Models and Annotation Tools for Comprehensive Affective Analysis		The University of Manchester
2023	Large Language Models Understand and Can Be Enhanced by Emotional Stimuli		Microsoft, Institute of Software, CAS
2023	Towards Emotion-Based Synthetic Consciousness: Using LLMs to Estimate Emotion Probability Vectors		Warwick University
2024	Enhancing Emotional Generation Capability of Large Language Models via Emotional Chain-of-Thought		School of Computer Science and Technology, Harbin Institute of Technology, Shenzhen
2024	Enhancing Empathetic Response Generation by Augmenting LLMs with Small-scale Empathetic Models		Anonymous Review
2023	SoulChat: Improving LLMs’ Empathy, Listening, and Comfort Abilities through Fine-tuning with Multi-turn Empathy Conversations		Guangdong Provincial Key Laboratory of Human Digital Twin, School of EE.
2022	Chain-of-Thought Prompting Elicits Reasoning in Large Language Models		Google Research, Brain Team
2023	METAAGENTS: SIMULATING INTERACTIONS OF HUMAN BEHAVIORS FOR LLM-BASED TASK-ORIENTED COORDINATION VIA COLLABORATIVE GENERATIVE AGENTS		University of Cambridge
2023	Meta-control of Dialogue Systems Using Large Language Models		-
2023	PLUG-AND-PLAY POLICY PLANNER FOR LARGE LANGUAGE MODEL POWERED DIALOGUE AGENTS		NUS
2023	DiagGPT: An LLM-based Dialogue System with Automatic Topic Management for Goal-Oriented Dialogue		Anonymous Review
2023	Prompt-Based Monte-Carlo Tree Search for Goal-oriented Dialogue Policy Planning		Department of Computer Science, Columbia University, New York, NY
2023	ZERO-SHOT GOAL-DIRECTED DIALOGUE VIA RL ON IMAGINED CONVERSATIONS		UC Berkeley
2023	Enhancing Large Language Model Induced Task-Oriented Dialogue Systems Through Look-Forward Motivated Goals		NUS, NTU, UCSB, UCL
2024	Bridging Human and AI Decision-Making with LLMs: The RAGADA Approach		Department of Computer Science and Engineering, Aalto University, Finland
2024	Personalized LLM Response Generation with Parameterized Memory Injection		Worcester Polytechnic Institute, Worcester, USA
2023	Large language models show human-like content biases in transmission chain experiments		Stanford University, Stanford, CA
2023	LLM-based Smart Reply (LSR): Enhancing Collaborative Performance with ChatGPT-mediated Smart Reply System		Clemson University, USA
2023	Translating Natural Language to Planning Goals with Large-Language Models		NUS

Datasets

Year	Title	Link
2024	Amazon Product Reviews	Amazon Product Reviews
2008	IEMOCAP	IEMOCAP
2017	DailyDialog	DailyDialog
2020	EmpatheticDialogues	EmpatheticDialogues
2022	PENS	PENS
2023	SoulChatCorpus	SoulChatCorpus
2023	MISC	MISC
2022	MultiESC	MultiESC
2022	GLHG	GLHG
2022	PsyQA	PsyQA
2023	efaqa	efaqa
2021	GSM8K	GSM8K
2021	CSQA	CSQA
2021	StrategyQA	StrategyQA
2022	SayCan	SayCan
2021	CraisglistBargain	CraigslistBargain
2022	CIMA	CIMA
2021	DialoGLUE	DialoGLUE
2021	PersuasionForGood	PersuasionForGood
2020	MultiWoZ 2.1	MultiWoZ 2.1
2021	AmazonQA	AmazonQA
2021	ALFRED-L	ALFRED-L

Open-Source Codes

Year	Paper	Code
2023	Emotional Intelligence of Large Language Models
2024	EmoLLMs: A Series of Emotional Large Language Models and Annotation Tools for Comprehensive Affective Analysis
2023	SoulChat: Improving LLMs’ Empathy, Listening, and Comfort Abilities through Fine-tuning with Multi-turn Empathy Conversations
2023	PLUG-AND-PLAY POLICY PLANNER FOR LARGE LANGUAGE MODEL POWERED DIALOGUE AGENTS
2023	Prompt-Based Monte-Carlo Tree Search for Goal-oriented Dialogue Policy Planning
2023	Enhancing Large Language Model Induced Task-Oriented Dialogue Systems Through Look-Forward Motivated Goals

Citation

If you found this project helpful, please kindly cite our work:

@misc{advikbibek24,
  author = {Advik Raj Basani, Bibekananda Kundu},
  title = {EET-LLM: A Emotional, Empathetic and Task-Oriented LLM},
  year = {2024},
  howpublished = {\url{https://github.com/BitC3t/EET-LLM}}
}

Execution and Running

You may run this framework locally via the following command: streamlit run frontend.py

Other scripts in this repo deal with the several other components of the framework.

Authors

This project was created by Advik Raj Basani ([email protected]), under the guidance of Dr. Bibekananda Kundu.

License

This code is under the GNU-GPL v3.0.