# SocialMaze: A Benchmark for Evaluating and Enhancing Social Reasoning in Large Language Models in Complex Social Environments

arXiv: https://arxiv.org/abs/2505.23713 · HTML: https://mlsanity.com/paper/2505.23713/

## Key facts

- Cited by 11 papers in the map (data through 14 Oct 2026).
- Placed in Reasoning benchmarks and evaluation, one of 17 directions in LLM reasoning.
- Builds on 15 of its 106 references that are in the map.
- First posted on arXiv 29 May 2025.

## Placed in

- LLM reasoning / [Reasoning benchmarks and evaluation](https://mlsanity.com/lr/reasoning-benchmarks-and-evaluation/)

## Builds on (most cited first, 15 of 15)

- [GPT-4o System Card](https://mlsanity.com/paper/2410.21276/) (2024)
- [OpenAI o1 System Card](https://mlsanity.com/paper/2412.16720/) (2024)
- [DeepSeek-V3 Technical Report](https://mlsanity.com/paper/2412.19437/) (2024)
- [Phi-4 Technical Report](https://mlsanity.com/paper/2412.08905/) (2024)
- [Exploring Large Language Models for Communication Games: An Empirical Study on Werewolf](https://mlsanity.com/paper/2309.04658/) (2023)
- [LLM as a Mastermind: A Survey of Strategic Reasoning with Large Language Models](https://mlsanity.com/paper/2404.01230/) (2024)
- [Werewolf Arena: A Case Study in LLM Evaluation via Social Deduction](https://mlsanity.com/paper/2407.13943/) (2024)
- [On the Trustworthiness of Generative Foundation Models: Guideline, Assessment, and Perspective](https://mlsanity.com/paper/2502.14296/) (2025)
- [Social Science Meets LLMs: How Reliable Are Large Language Models in Social Simulations?](https://mlsanity.com/paper/2410.23426/) (2024)
- [An Open Review of OpenReview: A Critical Analysis of the Machine Learning Conference Review Process](https://mlsanity.com/paper/2010.05137/) (2020)
- [A Large-Scale Simulation on Large Language Models for Decision-Making in Political Science](https://mlsanity.com/paper/2412.15291/) (2024)
- [Exploring Large Language Models for Word Games:Who is the Spy?](https://mlsanity.com/paper/2503.15235/) (2025)
- [Evaluate Bias without Manual Test Sets: A Concept Representation Perspective for LLMs](https://mlsanity.com/paper/2505.15524/) (2025)
- [The Open Review-Based (ORB) dataset: Towards Automatic Assessment of Scientific Papers and Experiment Proposals in High-Energy Physics](https://mlsanity.com/paper/2312.04576/) (2023)
- [DebateBench: A Challenging Long Context Reasoning Benchmark For Large Language Models](https://mlsanity.com/paper/2502.06279/) (2025)

## Built on by (most cited first, 11 of 11)

- [DyFlow: Dynamic Workflow Framework for Agentic Reasoning](https://mlsanity.com/paper/2509.26062/) (2025)
- [Collaborative Multi-Agent Scripts Generation for Enhancing Imperfect-Information Reasoning in Murder Mystery Games](https://mlsanity.com/paper/2604.11741/) (2026)
- [Inside the Unfair Judge: A Mechanistic Interpretability Account of LLM-as-Judge Bias](https://mlsanity.com/paper/2607.11871/) (2026)
- [Think Socially via Cognitive Reasoning](https://mlsanity.com/paper/2509.22546/) (2025)
- [Modeling the Mental World for Embodied AI: A Comprehensive Review](https://mlsanity.com/paper/2601.02378/) (2025)
- [GroupToM-Bench: Benchmarking Group Theory of Mind and Nonlinear Social Emergence in MLLMs](https://mlsanity.com/paper/2606.04184/) (2026)
- [SocialGrid: A Benchmark for Planning and Social Reasoning in Embodied Multi-Agent Systems](https://mlsanity.com/paper/2604.16022/) (2026)
- [PIVOTSBench: Evaluating Fine-Grained Interpersonal Relationship Reasoning in Multimodal Large Language Models](https://mlsanity.com/paper/2606.23092/) (2026)
- [Don' t Box Me In: Dynamic Cultural Adaptation and Cognitive Tracking for Social Understanding](https://mlsanity.com/paper/2608.22411/) (2026)
- [HEAR Who Said What: Unlocking Speaker-Attributed Reasoning via Counterfactual Voice Grounding](https://mlsanity.com/paper/2608.29120/) (2026)
- [GTA: Graph Theory Agent and Benchmark for Algorithmic Graph Reasoning with LLMs](https://mlsanity.com/paper/2609.12265/) (2026)

## Abstract (arXiv)

Large language models (LLMs) are increasingly deployed in socially grounded applications, where success requires interpreting context, inferring others' mental states, and reasoning about unreliable information. Yet existing benchmarks rarely evaluate these demands jointly in complex, evolving settings. We introduce SocialMaze, a benchmark that organizes six tasks across social deduction games, daily-life interactions, and digital community platforms along three descriptive design axes: deep reasoning, dynamic interaction, and information uncertainty. These axes characterize intended sources of task difficulty rather than latent, factor-analytic dimensions of model capability. Automated checks and human validation support data quality. Evaluations of twelve proprietary and open-weight LLMs show substantial variation in the use of evolving interaction histories; stronger chain-of-thought reasoners perform better on tasks requiring deeper inference, while uncertainty consistently degrades performance. Reasoning workflows help weaker short-chain-of-thought backbones but saturate on stronger reasoners. Finally, targeted fine-tuning on curated reasoning traces substantially improves structured social-reasoning tasks, whereas transfer to language-aggregation tasks remains statistically inconclusive. The project homepage is available at https://xzx34.github.io/socialmaze/.