Post-Doctoral Research Visit F/M Generating high-quality explanations of implicit arguments in controversial debates

The Inria centre at Université Côte d'Azur includes 42 research teams and 9 support services. The centre's staff (about 500 people) is made up of scientists of different nationalities, engineers, technicians and administrative staff. The teams are mainly located on the university campuses of Sophia Antipolis and Nice as well as Montpellier, in close collaboration with research and higher education laboratories and establishments (Université Côte d'Azur, CNRS, INRAE, INSERM ...), but also with the regiona economic players.

With a presence in the fields of computational neuroscience and biology, data science and modeling, software engineering and certification, as well as collaborative robotics, the Inria Centre at Université Côte d'Azur  is a major player in terms of scientific excellence through its results and collaborations at both European and international levels.

Motivation

Argumentation is a core mechanism of human reasoning, decision-making, and communication. In real-world debates, arguments are rarely fully explicit; they often rely on implicit knowledge, shared assumptions, or omitted premises. Recovering these missing elements is crucial for enabling machines to understand, interpret, and reason over natural language arguments.

This postdoctoral project focuses on the generation and evaluation of explanations of implicit arguments. The goal is not only to produce plausible explanations, but also to rigorously assess the quality of these explanations. For this purpose, we will extend and test in practice the methodology developed in “An Axiomatic Study of the Evaluation of Enthymeme Decoding in Weighted Structured Argumentation” (arXiv:2411.04555) by Victor David and Anthony Hunter. This framework provides a formal foundation for evaluating the quality of explanations based on multiple logical, structural, and semantical criteria.

Beyond the formal study, this postdoctoral project aims to bring our framework into real-world use by applying it to natural language arguments. We will evaluate the quality of generated explanations from real debates, combining formal criteria from arXiv:2411.04555 (adapted to natural language arguments) with NLP metrics such as BERTScore and repetition rate. If needed, we also plan to define new task-specific metrics better suited to this setting. In addition, we intend to carry out a human evaluation to assess how the generated explanations are perceived according to various criteria, such as clarity, informativeness, and plausibility, in order to support their interpretability and practical relevance.

Our long-term objective is to contribute to the understanding and mediation of complex debates, such as those involving environmental issues, where arguments span multiple dimensions (e.g., economic, environmental, moral, societal). In such contexts, making implicit reasoning steps explicit (and doing so in a structured and high-quality way) is essential to support constructive dialogue, critical thinking, and informed decision-making.

An Active Collaboration

Thanks to the collaboration between Inria Sophia Antipolis and University College London (UCL), and more specifically through the recently established Inria–UCL Associated Team EXPLAINER, a postdoctoral position may be funded (salary = 2 927€) based on the strength of the application. The selection will be made among the best applications submitted across all Inria associated teams. The results will be announced on July 4th.

Since October 2023, a strong collaboration has been in place between Anthony Hunter and Victor David, focusing on the evaluation and explanation of implicit arguments. The postdoc will further benefit from the involvement of Serena Villata and Elena Cabrio, who bring extensive expertise in natural language arguments and will support the project as part of the associated team.

Assignments & Collaboration

With the help of Victor David and Anthony Hunter, the recruited person will build upon existing symbolic and formal models of implicit arguments to design learning-based methods and evaluation metrics. The objective is not to propose new symbolic models, but rather to translate and operationalize these theoretical insights into computational approaches for explanation generation and evaluation. Victor and Anthony will provide close guidance to ensure the proper transfer and integration of formal knowledge.

In parallel, the postdoctoral researcher will work in close collaboration with Serena Villata and Elena Cabrio, experts in natural language processing (NLP) and argument mining, who will supervise the NLP components, dataset exploration, and empirical evaluation protocols. This interdisciplinary framework ensures that both formal rigor and empirical relevance are at the core of the project.

 The postdoctoral researcher will:

• Design methods for generating textual explanations of implicit arguments.
• Develop and apply evaluation frameworks to assess the quality of the explanations.
• Identify existing datasets related to controversial debates, such as environmental discourse.
• Participate in collaborative research and publications.

 The candidate must hold a PhD in Computer Science, with expertise in the following areas:

• Natural Language Processing (NLP).

• Argument Mining or Computational Argumentation.

• Knowledge Representation, Logic-based Reasoning, or Implicit Reasoning (optional but desirable).

• Experience in writing scientific publications and presenting research results.

Language: English.

• Subsidized meals
• Partial reimbursement of public transport costs
• Leave: 7 weeks of annual leave + 10 extra days off due to RTT (statutory reduction in working hours) + possibility of exceptional leave (sick children, moving home, etc.)
• Possibility of teleworking (after 6 months of employment) and flexible organization of working hours
• Professional equipment available (videoconferencing, loan of computer equipment, etc.)
• Social, cultural and sports events and activities
• Access to vocational training
• Social security coverage

Gross Salary: 2788 € per month