@inproceedings{zaghouani-2026-accountable,
    title = "Accountable Human-{AI} Deliberation with {LLM}s: Scaling Collective Intelligence through Symbiotic Scaffolding",
    author = "Zaghouani, Wajdi",
    editor = "Anastasiou, Lucas  and
      Boland, Katarina  and
      Liddo, Anna De  and
      Falk, Neele  and
      Hautli-Janisz, Annette  and
      Lapesa, Gabriella  and
      Romberg, Julia",
    booktitle = "Proceedings of The 2nd Workshop on Language-driven Deliberation Technology",
    month = may,
    year = "2026",
    address = "Mallorca, Spain",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/paragraph-normalization/2026.delite-1.2/",
    doi = "10.63317/37crtjgcefv3",
    pages = "7--17",
    abstract = "Large language models (LLMs) can support democratic deliberation at scales previously constrained by turn-taking and facilitation bandwidth. Recent work shows that LLM-generated group statements are often preferred over human-mediated outputs, while theoretical analyses argue that LLMs relax the simultaneity constraints limiting collective intelligence. Yet pure LLM mediation risks collapsing pluralism, over-optimizing for agreement, and undermining legitimacy when participants cannot contest how they are represented. We propose a symbiotic human-AI framework organized into three layers: observation and diversity amplification, facilitation with clause-level provenance, and human primacy for ratification. Our contributions include graded coverage, diversity, and erasure metrics with salience-aware weighting; a provenance pipeline combining cross-encoder similarity with causal knockout diagnostics; preference-conditioned trade-off control; equity-aware contestability workflows; adversarial robustness tests; and an evaluation protocol with ablation designs informed by evidence of LLM-as-judge limitations. The result is a testable blueprint for deliberation technology that scales collective intelligence while preserving agency and legitimacy."
}