Argumentative Human-AI Decision-Making: Toward AI Agents That Reason With Us, Not For Us
Organizations: New Mexico State University, Las Cruces, NM, USA · King’s College London, London, UK · Imperial College London, London, UK · Washington University in St. Louis, St. Louis, MO, USA
Abstract
Computational argumentation offers formal frameworks for transparent, verifiable reasoning but has traditionally been limited by its reliance on domain-specific information and extensive feature engineering. In contrast, LLMs excel at processing unstructured text, yet their opaque nature makes their reasoning difficult to evaluate and trust. We argue that the convergence of these fields will lay the foundation for a new paradigm: Argumentative Human-AI Decision-Making. We analyze how the synergy of argumentation framework mining, argumentation framework synthesis, and argumentative reasoning enables agents that do not just justify decisions, but engage in dialectical processes where decisions are contestable and revisable -- reasoning with humans rather than for them. This convergence of computational argumentation and LLMs is essential for human-aware, trustworthy AI in high-stakes domains.
Figures & tables
| Task | Description | Representa-tive Work |
|---|---|---|
| Argumentation Mining | Definition: Automatic identification and extraction of argumentative structures from text Enhancement: LLMs Eliminating need for manual features and structured inputs Subtasks : AF Extraction , Content Detection | ( Cabessa et al., 2025 ; Ruiz-Dolz et al., 2021 ; Gorur et al., 2025 ; Stahl et al., 2025 ; Rocha et al., 2024 ; Lan et al., 2024 ; Li et al., 2025 ; Bezou-Vrakatseli et al., 2025 ) |
| Argumentation Synthesis | Definition: Generation of new arguments, premises, claims, and summaries Enhancement: LLMs Enabling flexible, scalable generation without templates Subtasks : AF Generation , AF Summarization | ( Chen et al., 2024 ; Gray et al., 2025 ; Mouchel et al., 2025 ; Hu et al., 2024 ; Altemeyer et al., 2025 ; Roush et al., 2024 ; Li et al., 2024 ) |
| Argumentative Reasoning | Definition: Application of computational argumentation for sound decision-making Enhancement: Argumentation Providing formal semantics and transparent, contestable reasoning Subtasks : Claim Verification , Explainable Decision-Making | ( Yue et al., 2024 ; Si et al., 2024 ; Freedman et al., 2025 ; Ng et al., 2025 ; Zhu et al., 2025 ; Hong et al., 2024 ; Lin et al., 2024 ) |