# PerSo: Investigating Facets of LLMs' Social Intelligence in Real-World Persuasion Scenarios

arXiv: https://arxiv.org/abs/2610.20548 · HTML: https://mlsanity.com/paper/2610.20548/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 12 of its 22 references that are in the map.
- First posted on arXiv 14 Oct 2026.

## Builds on (most cited first, 12 of 12)

- [Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks](https://mlsanity.com/paper/1908.10084/) (2019)
- [SOTOPIA: Interactive Evaluation for Social Intelligence in Language Agents](https://mlsanity.com/paper/2310.11667/) (2023)
- [FANToM: A Benchmark for Stress-testing Machine Theory of Mind in Interactions](https://mlsanity.com/paper/2310.15421/) (2023)
- [NegotiationToM: A Benchmark for Stress-testing Machine Theory of Mind on Negotiation Surrounding](https://mlsanity.com/paper/2404.13627/) (2024)
- [SocialBench: Sociality Evaluation of Role-Playing Conversational Agents](https://mlsanity.com/paper/2403.13679/) (2024)
- [AgentSense: Benchmarking Social Intelligence of Language Agents through Interactive Scenarios](https://mlsanity.com/paper/2410.19346/) (2024)
- [COKE: A Cognitive Knowledge Graph for Machine Theory of Mind](https://mlsanity.com/paper/2305.05390/) (2023)
- [InterIntent: Investigating Social Intelligence of LLMs via Intention Understanding in an Interactive Game Context](https://mlsanity.com/paper/2406.12203/) (2024)
- [Towards Dynamic Theory of Mind: Evaluating LLM Adaptation to Temporal Evolution of Human States](https://mlsanity.com/paper/2505.17663/) (2025)
- [PersuasiveToM: A Benchmark for Evaluating Machine Theory of Mind in Persuasive Dialogues](https://mlsanity.com/paper/2502.21017/) (2025)
- [MOMENTS: A Comprehensive Multimodal Benchmark for Theory of Mind](https://mlsanity.com/paper/2507.04415/) (2025)
- [Social Intelligence Data Infrastructure: Structuring the Present and Navigating the Future](https://mlsanity.com/paper/2403.14659/) (2024)

## Abstract (arXiv)

Understanding the social intelligence of large language models (LLMs) is increasingly important for applications involving human-AI interaction and multi-agent environments. Social intelligence comprises two aspects: cognitive intelligence, the ability to infer others' mental states, and behavioral intelligence, the ability to select appropriate social actions based on these inferences to achieve social goals. Persuasion provides a natural testbed in which both aspects are simultaneously at play, yet prior benchmarks typically evaluate them in isolation using synthetic scenarios. We introduce PerSo, a benchmark that evaluates both aspects on real-world persuasion dialogues from Reddit's ChangeMyView community, using an LLM-in-the-loop pipeline to construct the evaluation tasks, validated through human annotation. Our evaluation of diverse LLMs shows that performance generally scales with model size but varies across dimensions. The strongest LLM slightly surpasses the best observed human performance on the behavioral task, while the best human participant remains ahead on the cognitive task.