Research

Language models now sit between us and much of what we read.

I want to understand whose perspectives these models reflect, how they decide what matters, and how easily their answers can be steered. Each of these quietly shapes what we come to believe. The papers below are my attempts to find out.

2027

01
  1. IP&M 2027

    Political Ideology Shifts in Large Language Models

    Pietro Bernardelle, Stefano Civelli, Leon Fröhling, Riccardo Lunardi, Kevin Roitero, and Gianluca Demartini

    Information Processing & Management, vol. 64, no. 2, article 105124, February 2027

2026

09
  1. arXiv 2026 · Preprint

    Cross-lingual Calibration of Pre-Generation Success Probes for Multilingual LLM Routing

    Andrea Paganelli, Stefano Civelli, Pietro Bernardelle, and Gianluca Demartini

    arXiv preprint arXiv:2610.06216

  2. CIKM 2026

    LLMs Encode Relevance as a Layer-Wise Cross-Lingual Signal

    Pietro Bernardelle, Samaneh Mohtadi, Stefano Civelli, Joel Mackenzie, and Gianluca Demartini

    Proceedings of the 35th International ACM Conference on Knowledge and Information Management (CIKM)

  3. CIKM 2026

    Persona Conditioning as an Assessor-Sensitivity Probe for LLM-Based IR Evaluation

    Samaneh Mohtadi, Pietro Bernardelle, Joel Mackenzie, and Gianluca Demartini

    Proceedings of the 35th International ACM Conference on Knowledge and Information Management (CIKM)

  4. GenAI4SG @ CIKM 2026

    Context Shapes LLMs Retrieval-Augmented Fact-Checking Effectiveness

    Pietro Bernardelle, Stefano Civelli, Kevin Roitero, and Gianluca Demartini

    Proceedings of the 1st Workshop on Generative AI for Social Good and Vulnerable Populations (GenAI4SG) at CIKM 2026

  5. ACL 2026 · Short Paper

    A Shared Geometry of Difficulty in Multilingual Language Models

    Stefano Civelli, Pietro Bernardelle, Nicolò Brunello, and Gianluca Demartini

    Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 2: Short Papers)

  6. ACM TIST 2026 · Special Issue on Risks and Unintended Harms of Generative AI Systems

    Ideology-Based LLMs for Content Moderation

    Stefano Civelli, Pietro Bernardelle, Nardiena A. Pratama, and Gianluca Demartini

    ACM Transactions on Intelligent Systems and Technology, published online April 2026

  7. NLPerspectives @ LREC 2026

    SubData: Bridging Heterogeneous Datasets to Enable Theory-Driven Evaluation of Political and Demographic Perspectives in LLMs

    Pietro Bernardelle, Leon Froehling, Stefano Civelli, and Gianluca Demartini

    Proceedings of the Fifth Edition of NLPerspectives

  8. ICWSM 2026

    Political Advertising on Facebook During the 2022 Australian Federal Election: A Social Identity Perspective

    Stefano Civelli, Pietro Bernardelle, Frank Mols, and Gianluca Demartini

    Proceedings of the International AAAI Conference on Web and Social Media, vol. 20, no. 1, pp. 563–577, May 2026

  9. ICWSM 2026

    Towards Detecting Persuasion on Social Media: From Model Development to Insights on Persuasion Strategies

    Elyas Meguellati, Stefano Civelli, Pietro Bernardelle, Shazia Sadiq, Irwin King, and Gianluca Demartini

    Proceedings of the International AAAI Conference on Web and Social Media, vol. 20, no. 1, pp. 1587–1608, May 2026

2025

02
  1. TheWebConf 2025 · Short Paper

    Mapping and Influencing the Political Ideology of Large Language Models using Synthetic Personas

    Pietro Bernardelle, Leon Fröhling, Stefano Civelli, Riccardo Lunardi, Kevin Roitero, and Gianluca Demartini

    Companion Proceedings of the ACM on Web Conference 2025

  2. TheWebConf 2025 · MM4SG Workshop · Best Paper Award

    The Impact of Persona-based Political Perspectives on Hateful Content Detection

    Stefano Civelli, Pietro Bernardelle, and Gianluca Demartini

    Companion Proceedings of the ACM on Web Conference 2025

2024

01
  1. SIGIR-AP 2024 · Short Paper

    Optimizing LLMs with Direct Preferences: A Data Efficiency Perspective

    Pietro Bernardelle, and Gianluca Demartini

    Proceedings of the 2024 Annual International ACM SIGIR Conference on Research and Development in Information Retrieval in the Asia Pacific Region