---
title: "SPIN-Bench: How Well Do LLMs Plan Strategically and Reason Socially?"
canonical_url: "https://www.modelscope.cn/papers/127426"
md_url: "https://www.modelscope.cn/papers/127426.md"
arxiv_id: 2503.12349
published: 2025-10-15
last_updated: 2025-10-15
authors:
  - "Jianzhu Yao"
  - "Kevin Wang"
  - "Ryan Hsieh"
  - "Haisu Zhou"
  - "Tianqing Zou"
  - "Zerui Cheng"
  - "Zhangyang Wang"
  - "Pramod Viswanath"
model_name: SPIN-Bench
model_developer: "Princeton University、The University of Texas at Austin、Sentient Foundation"
domain:
  - "人工智能"
  - "大语言模型评估"
  - "战略规划"
  - "社会推理"
  - "多智能体系统"
type:
  - "人工智能"
  - "大语言模型评估"
  - "战略规划"
  - "社会推理"
  - "多智能体系统"
  - "Artificial Intelligence"
arxiv_url: "https://arxiv.org/abs/2503.12349"
pdf_url: "https://arxiv.org/pdf/2503.12349"
code_link: "https://github.com/spinbench/spinbench"
---

# SPIN-Bench: How Well Do LLMs Plan Strategically and Reason Socially?

> Reasoning and strategic behavior in social interactions is a hallmark of intelligence. This form of reasoning is significantly more sophisticated than isolated planning or reasoning tasks in static settings (e.g., math problem solving). In this paper, we…

「SPIN-Bench: How Well Do LLMs Plan Strategically and Reason Socially?」是 ModelScope 魔搭社区收录的论文，arXiv 2503.12349，作者为 Jianzhu Yao, Kevin Wang, Ryan Hsieh et al.，发表于 2025-10-15，属于 人工智能、大语言模型评估、战略规划 领域。

- **ArXiv**: 2503.12349
- **Published**: 2025-10-15
- **Authors**: Jianzhu Yao, Kevin Wang, Ryan Hsieh, Haisu Zhou, Tianqing Zou, Zerui Cheng, Zhangyang Wang, Pramod Viswanath
- **Model**: SPIN-Bench
- **Developer**: Princeton University、The University of Texas at Austin、Sentient Foundation
- **Domain**: 人工智能, 大语言模型评估, 战略规划, 社会推理, 多智能体系统
- **ArXiv URL**: https://arxiv.org/abs/2503.12349
- **PDF**: https://arxiv.org/pdf/2503.12349
- **Code**: https://github.com/spinbench/spinbench

Source: https://www.modelscope.cn/papers/127426

---

> SPIN-Bench：大型语言模型在战略规划与社会推理方面的表现如何？

## 摘要

本文提出了 SPIN-Bench（Strategic Planning, Interaction, and Negotiation），一个统一的多领域评估基准框架，旨在全面衡量大型语言模型（LLMs）在战略规划和社会推理方面的能力。该基准将规划域定义语言（PDDL）任务、竞争性棋盘游戏、合作性卡牌游戏以及多智能体谈判场景整合在一个统一的框架中，系统地改变动作空间、状态复杂度和交互智能体数量。研究揭示了当前 LLMs 在长周期规划和复杂多智能体协调中的显著不足，并发现大规模社会交互会负面影响强规划模型的思维链连贯性。

## Abstract

Reasoning and strategic behavior in social interactions is a hallmark of intelligence. This form of reasoning is significantly more sophisticated than isolated planning or reasoning tasks in static settings (e.g., math problem solving). In this paper, we present Strategic Planning, Interaction, and Negotiation (SPIN-Bench), a new multi-domain evaluation designed to measure the intelligence of strategic planning and social reasoning. While many existing benchmarks focus on narrow planning or single-agent reasoning, SPIN-Bench combines classical PDDL tasks, competitive board games, cooperative card games, and multi-agent negotiation scenarios in one unified framework. The framework includes both a benchmark as well as an arena to simulate and evaluate the variety of social settings to test reasoning and strategic behavior of AI agents. We formulate the benchmark SPIN-Bench by systematically varying action spaces, state complexity, and the number of interacting agents to simulate a variety of social settings where success depends on not only methodical and step-wise decision making, but also conceptual inference of other (adversarial or cooperative) participants. Our experiments reveal that while contemporary LLMs handle basic fact retrieval and short-range planning reasonably well, they encounter significant performance bottlenecks in tasks requiring deep multi-hop reasoning over large state spaces and socially adept coordination under uncertainty. We envision SPIN-Bench as a catalyst for future research on robust multi-agent planning, social reasoning, and human--AI teaming. Project Website: https://spinbench.github.io/