# Make an Offer They Can't Refuse: Grounding Bayesian Persuasion in Real-World Dialogues without Pre-Commitment

arXiv: https://arxiv.org/abs/2510.13387 · HTML: https://mlsanity.com/paper/2510.13387/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 21 of its 60 references that are in the map.
- First posted on arXiv 15 Oct 2025.

## Builds on (most cited first, 21 of 21)

- [Qwen3 Technical Report](https://mlsanity.com/paper/2505.09388/) (2025)
- [Towards Understanding Sycophancy in Language Models](https://mlsanity.com/paper/2310.13548/) (2023)
- [Persuasion for Good: Towards a Personalized Persuasive Dialogue System for Social Good](https://mlsanity.com/paper/1906.06725/) (2019)
- [INSPIRED: Toward Sociable Recommendation Dialog Systems](https://mlsanity.com/paper/2009.14306/) (2020)
- [Steering Language Models with Game-Theoretic Solvers](https://mlsanity.com/paper/2402.01704/) (2024)
- [Optimal Rates and Efficient Algorithms for Online Bayesian Persuasion](https://mlsanity.com/paper/2303.01296/) (2023)
- [Incorporating Pragmatic Reasoning Communication into Emergent Language](https://mlsanity.com/paper/2006.04109/) (2020)
- [Towards Strategic Persuasion with Language Models](https://mlsanity.com/paper/2509.22989/) (2025)
- [Are the Values of LLMs Structurally Aligned with Humans? A Causal Perspective](https://mlsanity.com/paper/2501.00581/) (2024)
- [Efficient Model-agnostic Alignment via Bayesian Persuasion](https://mlsanity.com/paper/2405.18718/) (2024)
- [Persuasion Should be Double-Blind: A Multi-Domain Dialogue Dataset With Faithfulness Based on Causal Theory of Mind](https://mlsanity.com/paper/2502.21297/) (2025)
- [Polynomial-Time Optimal Equilibria with a Mediator in Extensive-Form Games](https://mlsanity.com/paper/2206.15395/) (2022)
- [Rationality-Robust Information Design: Bayesian Persuasion under Quantal Response](https://mlsanity.com/paper/2207.08253/) (2022)
- [Roadmap on Incentive Compatibility for AI Alignment and Governance in Sociotechnical Systems](https://mlsanity.com/paper/2402.12907/) (2024)
- [Bayesian Persuasion Meets Mechanism Design: Going Beyond Intractability with Type Reporting](https://mlsanity.com/paper/2202.00605/) (2022)
- [Online Bayesian Persuasion Without a Clue](https://mlsanity.com/paper/2411.06141/) (2024)
- [Verbalized Bayesian Persuasion](https://mlsanity.com/paper/2502.01587/) (2025)
- [When and Why is Persuasion Hard? A Computational Complexity Result](https://mlsanity.com/paper/2408.07923/) (2024)
- [Persuasion Games using Large Language Models](https://mlsanity.com/paper/2408.15879/) (2024)
- [PoliCon: Evaluating LLMs on Achieving Diverse Political Consensus Objectives](https://mlsanity.com/paper/2505.19558/) (2025)
- [Bayesian Persuasion with Externalities: Exploiting Agent Types](https://mlsanity.com/paper/2412.12859/) (2024)

## Built on by (most cited first, 1 of 1)

- [Credibility Governance: A Social Mechanism for Collective Self-Correction under Weak Truth Signals](https://mlsanity.com/paper/2603.02640/) (2026)

## Abstract (arXiv)

Large language models (LLMs) still struggle with strategic persuasion, largely because existing approaches either neglect information asymmetry or rely on unrealistic pre-commitment assumptions. We introduce a type-induced commitment-communication mechanism that grounds Bayesian Persuasion (BP) in natural language dialogue without pre-commitment: the persuader narrates their potential types (e.g., honest vs. dishonest) to dynamically construct an information schema, enabling the persuadee to perform Bayesian belief updates within the conversation itself. We implement two variants: Semi-Formal-Natural-Language (SFNL) and Fully-Natural-Language (FNL), evaluating them against strong baselines across multiple LLMs and human judges. BP strategies consistently outperform baselines: SFNL excels in logical credibility, while FNL shows superior robustness and emotional resonance. We verify that gains stem from genuine Bayesian reasoning rather than superficial formatting, and we further show that supervised fine-tuning enables small models to match the persuasive performance of much larger ones.