PhD Student Vacancy for Explainable Artificial Intelligence
About the Project
This project, Design, Development, and Evaluation of an Intelligent Meta-Agent Capable of Negotiating in Natural Language, is a 36-month initiative funded by The Scientific and Technological Research Council of Turkey (TÜBİTAK) under the 1001 project program.
The core of this project is to create a sophisticated negotiation agent that is capable of negotiating in natural language by fusing traditional, strategic negotiation mechanisms with the power of Large Language Models (LLMs). The goals of the project can be summarized as follows:
- Natural Language Fluency: Using LLMs to understand, process, and generate human-like dialogue, arguments, and responses.
- Strategic Adaptivity: Leveraging the advanced reasoning capabilities of LLMs, combined with proven negotiation algorithms to dynamically analyze human sentiment, model opponent preferences, and adjust the underlying negotiation strategy in real-time.
- Evaluation: Developing a dedicated platform for negotiations in natural language to conduct our evaluations, using both simulations and user experiments to validate the agent's performance.
Job Description
We are looking for a PhD or MSc. student to work together on the following fields:
- Argumentation
- Opponent Modelling
- Negotiation Strategies
- Natural Language Processing
- Large Language Models
- Multi-Agent Systems
We expect you to actively participate in the following tasks:
- Conducting high-quality research within the project
- Presentation of research findings at workshops and conferences
- Publication of scientific papers in peer-reviewed international journals
- Collaborate strongly with all peers
Requirements
- Specialization in artificial intelligence or related field
- Written and oral fluency in English
- Minimum of 3.00 GPA in undergraduate and minimum of 3.50 GPA in master’s degree
- An intermediate understanding of Software Engineering ( e.g., Python, Github )
- An intermediate understanding of Machine Learning (e.g., Hugging Face, PyTorch)
Nice to Have’s:
- Prior publications
- Familiarity with automated negotiation or multi-agent systems
- Agent Organization Frameworks (e.g., LangGraph, LangChain)
Please contact Reyhan Aydogan via email.