Reliable Conversational Domain-specific
Data Exploration and Analysis

Doctoral Candidates

The ARMADA Doctoral Network brings together 15 highly skilled Early Stage Researchers from diverse backgrounds across Europe. These doctoral candidates are specializing in conversational AI, large language models, and reliable data exploration systems.

Program Details

Total Candidates 15

Partner Institutions 8

Countries Represented 7

Program Duration 4 years

Funding: Horizon Europe - MSCA Doctoral Networks Grant Agreement No. 101168951
and Swiss State Secretariat for Education, Research and Innovation (SERI) under SBFI No. 24.00005

WP1: Efficiency

Francesca Del Gaudio

Francesca Del Gaudio

Université Paris Cité France Italian

Fast And Accurate Similarity Search in High-Dimensional Vectors

Paolo Serafino Mastroianni

Paolo Serafino Mastroianni

KTH Royal Institute of Technology Sweden Italian

Learning-augmented algorithms for efficient optimization and retrieval

Konstantinos Nikoletos

Konstantinos Nikoletos

Utrecht University Netherlands Greek

Data Reduction

WP2: Grounding

Thushari Pahalage

Thushari Pahalage

University of Verona Italy Sri Lankan

Neuro-symbolic knowledge-aware analytical processing

Clara Liang

Clara Liang

University of Verona Italy Chinese

Analysis of bias in training materials for chatbots/LLM via linguistic corpora analysis

Gabriele Carrino

Gabriele Carrino

IBM Research Zurich Switzerland Italian

Self-alignment with Domain Knowledge

WP3: Explainability

Amanda Le

Amanda Le

CNRS, University Grenoble Alpes France Danish

Graph data model for explanations

Tommaso Dolci

Tommaso Dolci

TU Wien Austria Italian

Knowledge-Graph driven Factuality and Explainability

Dina Fettache

Dina Fettache

CNRS, University Grenoble Alpes France Algerian

Explainability for Conversation Pathways

WP4: Soundness

Ankita Maity

Ankita Maity

Aarhus University Denmark Indian

Robustness of estimation and confidence

Gokul Karthik Kumar

Gokul Karthik Kumar

IBM Research Zurich Switzerland Indian

Offline Reinforcement Learning and Reward Augmented Decoding

Vittorio La Ferla

Vittorio La Ferla

ZHAW – Zurich University of Applied Science Switzerland Italian

User-specific Reliable Knowledge Discovery for Effective Personalized Data Exploration

WP5: Guidance

Yuxuan Jiang

Yuxuan Jiang

Athena Research Center Greece Chinese

Diverse SQL query generation

Yifei Dong

Yifei Dong

Athena Research Center Greece Chinese

User Guidance in Multi-turn Conversation

Ibukun David Babatunde

Ibukun David Babatunde

Aarhus University Denmark Nigerian

Reliable Conversational QA over Knowledge Graphs via Query Reformulation

For more information about the program, visit the main project page or contact Matteo Lissandrini at project.armada@ateneo.univr.it