arXiv AI September 10, 2026 By Hwang Yechan, Bae Sangjun, Kim Jeongmo, Bang Sangwoo, Han Seungyul

MARBO: Relational Belief Grounding for LLM Agents in Social Deduction Games

The paper introduces MARBO, a framework that grounds large language model agents’ actions and speech in relational beliefs about hidden roles and team alignments in social deduction games. By providing preference feedback only when behaviors are supported by reliable relational beliefs and lead to favorable outcomes, MARBO encourages consistent learning under uncertainty. Experiments demonstrate that compact LLM agents using MARBO consistently outperform existing baselines in representative social deduction games.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv AI.

arXiv AI

Jun 4

SMAC-Talk: A Natural Language Extension of the StarCraft Multi-Agent Challenge for Large Language Models

arXiv:2606. 04202v1 Announce Type: new Abstract: As LLMs become more widely deployed, they are increasingly expected to work alongside other AI agents rather than operating in isolation.

By Joel Sol, Homayoun Najjaran

llms agents benchmarks

arXiv AI

Jun 2

TriAlign: Towards Universal Truth Consistency in Personalized LLM Alignment

arXiv:2606. 01755v1 Announce Type: new Abstract: Personalized large language models adapt responses to users' preferences and social attributes, but can introduce substantial universal truth inconsistencies across social groups, where some groups systematically receive less accurate responses on objective tasks.

By Thi-Nhung Nguyen, Linhao Luo, Rollin Omari, Junae Kim, Thuy-Trang Vu, Dinh Phung

llms agents reinforcement-learning benchmarks safety

arXiv Machine Learning

Jun 5

CollabBench: Benchmarking and Unleashing Collaborative Ability of LLMs with Diverse Players via Proactive Engagement

arXiv:2606. 05793v1 Announce Type: cross Abstract: While LLM-based agents excel at individual tasks, effective collaboration with realistic human partners remains challenging.

By Hong Qian, Yuanhao Liu, Zihan Zhou, Zongbao Zhang, Hanjie Ge, Haotian Shi, Liang Dou, Xiangfeng Wang, Jingwen Yang, Aimin Zhou

llms agents benchmarks

arXiv AI

Sep 17

Clueing up LLMs with Tool-Augmented Deductive Reasoning

The paper introduces a text-based, multi-agent version of the board game Clue to test multi-step deductive reasoning in large language models (LLMs). Six LLM-based agents (GPT‑4o‑mini and Gemini‑2.5‑Flash) play turn‑based games, and a tool‑augmented approach uses a structured possibility matrix to convert implicit game state into explicit remaining possibilities, thereby offloading memory and deductive constraints from the agents. The study compares this tool‑augmented method against a baseline to assess its impact on reasoning quality and task success in a strategic reasoning environment.

By Rebecca Ansell, Autumn Toney-Wails

llms agents

arXiv AI

Jul 14

Multi-Agent LLMs Fail to Explore Each Other

arXiv:2607. 11250v1 Announce Type: cross Abstract: Exploration is essential for reliable autonomy in multi-agent systems, yet it remains unclear whether large language model (LLM) agents can explore effectively when interacting with one another.

By Hyeong Kyu Choi, Jiatong Li, Wendi Li, Xin Eric Wang, Sharon Li

llms agents

arXiv AI

Aug 12

Multiplayer Nash Preference Optimization

arXiv:2509. 23102v4 Announce Type: replace Abstract: Reinforcement learning from human feedback (RLHF) has emerged as the standard paradigm for aligning large language models with human preferences.

By Fang Wu, Xu Huang, Weihao Xuan, Zhiwei Zhang, Yijia Xiao, Guancheng Wan, Xiaomin Li, Bing Hu, Peng Xia, Jure Leskovec, Yejin Choi

llms reinforcement-learning benchmarks safety

One email a morning, machine-written

Related stories

SMAC-Talk: A Natural Language Extension of the StarCraft Multi-Agent Challenge for Large Language Models

TriAlign: Towards Universal Truth Consistency in Personalized LLM Alignment

CollabBench: Benchmarking and Unleashing Collaborative Ability of LLMs with Diverse Players via Proactive Engagement

Clueing up LLMs with Tool-Augmented Deductive Reasoning

Multi-Agent LLMs Fail to Explore Each Other

Multiplayer Nash Preference Optimization