arXiv AI

Working with Agentic `Teammates': When a New Organizational Actor Collides with the Human Ecosystem of Work

The paper reports an in‑situ qualitative study of a persistent, proactive AI teammate deployed across multiple teams in a large technology company. It finds that the human‑agent workplace is in flux, with breakdowns and negotiations emerging around tacit workflow rules, the relational boundaries of the non‑human actor, and the redistribution of trust and human agency. These micro‑negotiations are used to propose a new research, design, and organizational agenda that seeks to preserve human agency when sharing workspaces with non‑human actors.

arXiv Computation and Language
Oct 15

Evolve Together, Fall into Place: How Social Scaffolding Shapes Self-Evolving Multi-Agent Systems

The paper introduces AGORA, a framework that independently varies team composition and interdependence structures to study how social scaffolding influences the evolution of large language model agents. Experiments across web search, codebase understanding, and document QA show that appropriate scaffolding can outperform a strong agent evolved alone by up to 25 points, primarily by empowering the most capable member to lead while others step back. The findings suggest that designing a self‑evolving multi‑agent system requires careful consideration of its social scaffolding to shape learning dynamics and leadership emergence.

By Seojin Hwang, Youngmin Choi, Hwanhee Lee
arXiv AI
Oct 16

Kolab: Benchmarking Human-Agent Collaboration in Long-Horizon Workflows

Kolab is a new dataset comprising over 115 hours of screen recordings from 50 professionals across 40 occupations who collaborate with AI agents on realistic, long‑horizon tasks. The dataset includes participants’ judgments of collaboration quality and output, revealing varied coordination patterns and the need for agents to anticipate how experts will use or modify their contributions. Analysis shows current state‑of‑the‑art models overestimate collaboration success and struggle to differentiate high‑ from low‑quality interactions, indicating a gap in understanding long‑horizon human‑agent teamwork.

By Shirley Anugrah Hayati, Patricia Anugrah Setiani, Ruizi Wang, Zhouyi Li, Khanh Chi Le, Young-Jun Lee, Dongyeop Kang
arXiv AI
Oct 16

One Is Not Enough: How People Use Multiple AI Models in Everyday Life

The article reports that people increasingly use multiple multimodal large language models (MLLMs) at the same time, choosing each model for its perceived strengths. This practice creates coordination challenges such as adapting prompts, calibrating trust, and managing separate conversation histories. A diary study and interviews with ten participants show that users build primary and secondary hierarchies among models, develop switching patterns based on task aggregation, and adjust effort, latency, and output credibility to manage multi‑MLLM workflows.

By Seunghwa Pyo, Donggun Lee, Jungwoo Rhee, Soobin Park, Youn-kyung Lim
arXiv AI
Aug 18

Position: AI Agents in Scientific Teams Should Be Studied as Human-Agent Systems

arXiv:2608. 14667v1 Announce Type: new Abstract: Large language model-based agents are increasingly deployed as collaborators in scientific discovery yet most current work focuses on the autonomous capabilities of "AI Scientists".

By Patrick Emami, Sameera Horawalavithana, Truc Nguyen, Gihan Panapitiya, Bruno Jacob, Siddhisanket Raskar, Saumya Sinha, Jared D. Willard, Andrew Glaws, Nithin Somasekharan, Ling Yue, Brian Lu, Shaowu Pan, Jason Eisner
arXiv AI
Sep 11

Pairit: A Platform for Live Experiments on Human-AI Collaboration

Pairit is an online platform that enables researchers to design, test, and deploy live experiments on human-AI collaboration. Using a single YAML configuration file, users can specify an executable experiment graph—including pages, routing, randomization, matchmaking, chat, shared workspaces, server-hosted agents, surveys, timers, and custom HTML components—and combine any number of humans and AI agents in real-time sessions. The platform has been validated through multiple live deployments, including peer-reviewed studies, and captures high-resolution process traces of communication, negotiation, and collaborative work in human-AI dyads.

By Harang Ju, Sinan Aral
arXiv AI
Sep 30

Human-AI Collaboration: From Paradoxes to Patterns

The paper demonstrates that humans and AI systems achieve better performance when collaborating rather than working alone. It investigates how two design dimensions—autonomy and initiative—shape collaboration patterns, using a paradox perspective to uncover internal tensions and map underlying paradoxes. From this analysis, the authors derive four distinct human‑AI collaboration patterns: Instruction, Delegation, Assistance, and Co‑creation.

By Michael Weiss
arXiv AI
Jun 6

Humans' ALMANAC: A Human Collaboration Dataset of Action-Level Mental Model Annotations for Agent Collaboration

arXiv:2606. 06388v1 Announce Type: new Abstract: Recent advances in LLM agents have enabled complex cognitive capabilities, such as multi-step reasoning, planning, and tool use, that increasingly position these agents as human collaborators.

By Jiaju Chen, Yuxuan Lu, Jiayi Su, Chaoran Chen, Songlin Xiao, Zheng Zhang, Yun Wang, Yunyao Li, Jian Zhao, Tongshuang Wu, Toby Jia-Jun Li, Dakuo Wang, Bingsheng Yao
arXiv AI
Oct 31

EvoTeam: Multi-Agent Collaboration via Evolving Teams

EvoTeam is a collaboration‑centric evolution framework for large‑language‑model based multi‑agent systems. It jointly evolves team organization, interaction rules, and individual agent skills using execution experience, while maintaining a library of reusable teams and guiding collaboration through team‑specific interaction rules. Experiments on four benchmarks show its effectiveness, and the code is publicly available on GitHub.

By Qinuo Cao, Xin He, Dianxuan Fu, Lizhi Zhang, Zhaoqing Xing, Jiapu Wang, Xin Wang, Na Zou, Yi Chang, Qinggang Zhang