***Post-doc position offer 18 months: LLM-based Multicriteria Decision Making**
**Offer** Large Language Models (LLMs) are increasingly employed in a wide range of decision making tasks. However, the opacity of their internal reasoning makes it difficult to validate or interpret their outputs, and the need for interpretability becomes especially critical in high-stakes settings. Our aim is to examine the decision-making capabilities of LLMs through Multicriteria Decision Making Methodologies (MCDM) and decision making under uncertainty approaches (Doe and Liu, 2023;Park et al., 2025;Zhiguang et al., 2026). We will focus in particular on enhancing LLMs decision making capabilities through criteria generation and dedicated reasoning processes while ensuring the reliability, interpretability and generalizability of their results. We will target different MCDM methods while considering how uncertainty and imprecision impact LLM-based MCDM.
**Applicant Profil**
-- Salary: €3,000 gross per month -- Expected start date: 1 February 2027 -- Duration: 18 months, full-time -- Status: Postdoctoral researcher
**Supervision** The successful candidate will undertake the post-doctoral research at the IRIT Laboratory https://www.irit.fr/en/home/, under the supervision of Prof. Pascale Zaraté (Toulouse Capitole University), Prof. Farah Benamara (University of Toulouse) and Dr Pierre Pomeret (Toulouse Capitole University – Rodez University Institute of Technology). The successful candidate will be based at the IRIT. The work will be carried out in collaboration with the AERODS https://www.ut-capitole.fr/home/research/teams-and-structures/aerods-centre-of-excellence-aerospace-defence-and-social-sciences Centre of Excellence – University of Toulouse Capitole.
**Required Skills** -- A PhD in Computer Science, -- Previous research experience in Large Language Models (LLMs) and/or multi-criteria decision-making (MCDM) methods. A solid kill in deep learning is required. -- Very good publication records. -- Strong writing and organizational skills, the ability to formulate and lead scientific projects, autonomy.
**Application** The application must be submitted by 30 November 2026 at the latest. It must include: a PhD degree certificate (or a proof of PhD degree by December 2026); list of publications highlighting one or two significant papers; letters of recommendation, if possible; and a detailed CV.
Applications must be sent as a single PDF document to: pascale.zarate@ut-capitole.fr, farah.benamara@irit.fr, pierre.pomeret@ut-capitole.fr
**References** -- Han Zhiguang, Farah Benamara, Pascale Zaraté (2026). Integrating the Analytic Hierarchy Process with Large Language Models for Transparent Multi-Criteria Decision-Making. arXiv preprint arXiv:2609.16779 -- Jane Doe and Han Liu. 2023. Can chatgpt serve as a multi-criteria decision maker? Proceedings of the 2023 Conference on Decision Science -- Haeun Park, Hyunjoo Oh, Feng Gao, and Ohbyung Kwon. 2025. Enhancing analytic hierarchy process modelling under uncertainty with fine-tuning LLM. Expert Systems, 42(6):e70051