CodeCRDT: Observation-Driven Coordination for Multi-Agent LLM Code Generation

2025-10-18 · Sergey Pugachev arxiv

Multi-agent LLM systems fail to realize parallel speedups due to costly coordination. We present CodeCRDT, an observation-driven coordination pattern where agents coordinate by monitoring a shared state with observable updates and deterministic convergence, rather than explicit message passing. Using Conflict-Free Replicated Data Types (CRDTs), CodeCRDT enables lock-free, conflict-free concurrent code generation with strong eventual consistency. Evaluation across 600 trials (6 tasks, 50 runs per mode) shows both benefits and trade-offs: up to 21.1% speedup on some tasks, up to 39.4% slowdown on others, and 100% convergence with zero merge failures. The study formalizes observation-driven coordination for stochastic LLM agents, revealing semantic conflict rates (5-10%) and quality-performance tradeoffs, and provides empirical characterization of when parallel coordination succeeds versus fails based on task structure.

📄 PDF Abstract BibTeX arXiv:2510.18893

Code (0)

등록된 구현이 없습니다.

Tasks

Code Generation

Coordination-driven learning in multi-agent problem spaces

2018-09-13 · Sean L. Barton, Nicholas R. Waytowich, Derrik E. Asher

We discuss the role of coordination as a direct learning objective in multi-agent reinforcement learning (MARL) domains. To this end, we present a novel means of quantifying coordination in multi-agent systems, and discu…

Multi-agent Reinforcement Learning reinforcement-learning Reinforcement Learning Reinforcement Learning (RL)

When Does Hierarchy Help? Benchmarking Agent Coordination in Event-Driven Industrial Scheduling

2026-05-13 · Ziqi Wang, Yuhao Yang, Zhiwei Ling, Wenzhuo Qian 외 arxiv

Recent advances in agent and multi-agent systems have shown strong performance on tool use, reasoning, and collaborative tasks. However, existing benchmarks mostly evaluate task completion in weakly coupled environments,…

Decision Making

Improving Coordination in Small-Scale Multi-Agent Deep Reinforcement Learning through Memory-driven Communication

2019-01-12 · Emanuele Pesce, Giovanni Montana

Deep reinforcement learning algorithms have recently been used to train multiple interacting agents in a centralised manner whilst keeping their execution decentralised. When the agents can only acquire partial observati…

Deep Reinforcement Learning Reinforcement Learning Reinforcement Learning (RL)

CoCoBench: A Cooperative Coordination Benchmark for Embodied Multi-Agent Task Planning

2026-08-28 · Yang Chen, Ye-Xin Xie, Lirong Che, Danyang Peng 외 arxiv

Agent systems powered by multimodal large language models (MLLMs) have advanced rapidly in recent years, yet existing embodied-agent benchmarks still lack fine-grained diagnostics for multi-agent coordination. Most bench…

Goal-Oriented Multi-Agent Reinforcement Learning for Decentralized Agent Teams

2025-11-15 · Hung Du, Hy Nguyen, Srikanth Thudumu, Rajesh Vasa 외 arxiv

Connected and autonomous vehicles across land, water, and air must often operate in dynamic, unpredictable environments with limited communication, no centralized control, and partial observability. These real-world cons…

Multi-agent Reinforcement Learning Autonomous Vehicles