paper-with-me

Papers

SocialGrid: A Benchmark for Planning and Social Reasoning in Embodied Multi-Agent Systems

2026-04-17 · Hikaru Shindo, Hanzhao Lin, Lukas Helff, Patrick Schramowski, Kristian Kersting arxiv

As Large Language Models (LLMs) transition from text processors to autonomous agents, evaluating their social reasoning in embodied multi-agent settings becomes critical. We introduce SocialGrid, an embodied multi-agent environment inspired by Among Us that evaluates LLM agents on planning, task execution, and social reasoning. Our evaluations reveal that even the strongest open model (GPT-OSS-120B) achieves below 60% accuracy in task completion and planning, with agents getting stuck in repetitive behaviors or failing to navigate basic obstacles. Since poor navigation confounds evaluation of social intelligence, SocialGrid offers an optional Planning Oracle to isolate social reasoning from planning deficits. While planning assistance improves task completion, social reasoning remains a bottleneck: agents fail to detect deception at near-random chance regardless of scale, relying on shallow heuristics rather than accumulating behavioral evidence. SocialGrid provides automatic failure analysis and fine-grained metrics, enabling developers to diagnose and improve their agents. We also establish a competitive leaderboard using Elo ratings from adversarial league play.

📄 PDF Abstract BibTeX arXiv:2604.16022

Code (0)

등록된 구현이 없습니다.

Similar Papers 제목 키워드 기반

Infer Human's Intentions Before Following Natural Language Instructions

2024-09-26 · Yanming Wan, Yue Wu, Yiping Wang, Jiayuan Mao 외

For AI agents to be helpful to humans, they should be able to follow natural language instructions to complete everyday cooperative tasks in human environments. However, real human instructions inherently possess ambigui…

Instruction Following

Constrained Human-AI Cooperation: An Inclusive Embodied Social Intelligence Challenge

2024-11-04 · Weihua Du, Qiushi Lyu, Jiaming Shan, Zhenting Qi 외

We introduce Constrained Human-AI Cooperation (CHAIC), an inclusive embodied social intelligence challenge designed to test social perception and cooperation in embodied agents. In CHAIC, the goal is for an embodied agen…

RynnBrain: Open Embodied Foundation Models

2026-02-13 · Ronghao Dang, Jiayan Guo, Bohan Hou, Sicong Leng 외 arxiv

Despite rapid progress in multimodal foundation models, embodied intelligence community still lacks a unified, physically grounded foundation model that integrates perception, reasoning, and planning within real-world sp…

Spatial Reasoning

Embodied AI Agents: Modeling the World

2025-06-27 · Pascale Fung, Yoram Bachrach, Asli Celikyilmaz, Kamalika Chaudhuri 외

This paper describes our research on AI agents embodied in visual, virtual or physical forms, enabling them to interact with both users and their environments. These agents, which include virtual avatars, wearable device…

Human Agent Collaboration

NormAct: A Benchmark for Hidden Social Norm Compliance in Embodied Planning

2026-06-26 · Shiyun Zhao, Xinwei Song, Tianyu Guo, Xiaomeng Gao 외 arxiv

Multimodal large language models (MLLMs) are increasingly deployed as embodied planners in egocentric environments, where task success requires not only achieving instructed goals but also acting in socially appropriate …