MOSAIC: A Universal Agent-Level Interface for Cross-Paradigm Agent Mixing and Human-AI Collaboration

arXiv 2603.01260 ↗First posted 1 Mar 2026cs.LG cs.AIcited by 0

On the map: LLM agents and tool use → Reinforcement learning for LLM agents

Abstract (arXiv)

Existing infrastructure cannot deploy agents from different decision-making paradigms within the same environment, making fair cross-paradigm comparison under identical conditions impossible. We present MOSAIC, an open-source platform that enables heterogeneous agents (RL policies, LLMs, VLMs, and human operators) to act within shared reinforcement learning environments in ad-hoc team settings with reproducible results. MOSAIC introduces three contributions. (i) IPC-based worker protocol that wraps native and third-party frameworks as isolated subprocess workers, each executing its own training and inference logic unmodified and communicating through a versioned inter-process protocol. (ii) An operator abstraction that forms an agent-level interface by mapping workers to agent slots: each operator, regardless of whether it is backed by an RL policy, an LLM, or a human, conforms to a minimal universal interface. (iii) A deterministic cross-paradigm evaluation framework with two complementary modes: a manual mode that advances up to N operators in lock-step under shared seeds for fine-grained visual inspection of behavioural differences; and a script mode that drives automated, long-running evaluation via declarative Python scripts for reproducible experiments. Our documentation is released at: https://mosaic-platform.readthedocs.io.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Placed in Reinforcement learning for LLM agents, one of 16 directions in LLM agents and tool use. Builds on 20 of its 26 references that are in the map. First posted on arXiv 1 Mar 2026.

Builds on

20 of 26 references.

Closest ancestors · 2

Same direction, newest first

Other references · 18

2019On the Utility of Learning about Humans for Human-AI Coordinationcited by 345
2020PettingZoo: Gym for Multi-Agent Reinforcement Learningcited by 209
2021Tianshou: a Highly Modularized Deep Reinforcement Learning Librarycited by 139
2020Sample Factory: Egocentric 3D Control from Pixels at 100000 FPS with Asynchronous Reinforcement Learningcited by 84
2023TorchRL: A data-driven decision-making library for PyTorchcited by 59
2023JaxMARL: Multi-Agent RL Environments and Algorithms in JAXcited by 59
2022EnvPool: A Highly Parallel Reinforcement Learning Environment Execution Enginecited by 58
2025lmgame-Bench: How Good are LLMs at Playing Games?cited by 54
2022A Survey of Ad Hoc Teamwork Researchcited by 41
2025Collab-Overcooked: Benchmarking and Evaluating Large Language Models as Collaborative Agentscited by 39
2023BenchMARL: Benchmarking Multi-Agent Reinforcement Learningcited by 31
2023ZSC-Eval: An Evaluation Toolkit and Benchmark for Multi-agent Zero-shot Coordinationcited by 23

6 more

2024Evaluating Large Language Models with Grid-Based Game Competitions: An Extensible LLM Benchmark and Leaderboardcited by 22
2024GAMEBoT: Transparent Assessment of LLM Reasoning in Gamescited by 16
2024N-Agent Ad Hoc Teamworkcited by 15
2022A General Learning Framework for Open Ad Hoc Teamwork Using Graph-based Policy Learningcited by 12
2023XuanCe: A Comprehensive and Unified Deep Reinforcement Learning Librarycited by 5
2026BotzoneBench: Scalable LLM Evaluation via Graded AI Anchorscited by 2

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way