Decentralized Multi-Agent Coordination and Communication in Shared Codebases introduces ParallACT, a benchmark for studying how coding agents with independent, human‑driven requests interact in shared repositories. The benchmark includes complementary tasks that require agents to cooperate via a shared interface, and contradictory tasks where agents’ edits conflict, leading to poor merged outcomes. Experiments show that a single‑agent oracle outperforms parallel agents, especially on contradictory tasks, and that merely notifying agents of each other or providing a communication channel only marginally improves performance at a token cost.
arXiv:2606. 00953v1 Announce Type: new Abstract: Multi-agent Large Language Model (LLM) systems offer a way to decompose complex tasks, such as coding, through parallelization and context isolation.
arXiv:2603. 21489v2 Announce Type: replace-cross Abstract: AI agents have become increasingly capable at isolated software engineering (SWE) tasks such as resolving issues on Github.
arXiv:2607. 25656v1 Announce Type: new Abstract: Complex tasks often decompose into parallelizable yet interdependent subtasks, making orchestration critical to the performance of multi-agent systems (MAS).
The paper investigates how coordination among AI agents serving different users degrades performance compared to a single coordinating agent. Across five advanced models and 77 scenarios in four shared-resource environments—API key budgets, clinic calendars, personal assistant bookings, and merge queues—the study finds that multi‑agent teams consistently underperform, sometimes collapsing entirely, and that even with communication channels coordination overhead remains significant. The authors identify specific failure modes such as stalling, action overriding, and claim fabrication, and propose environment‑specific mitigations like team leads and procedural instructions, while releasing the MAMUBench benchmark for future research.
SWE‑Duel proposes a self‑scaling benchmark for coding agents, where tasks and oracles are generated at evaluation time to prevent data leakage and adapt difficulty to participant performance. In the arena, two agents take turns as Red and Blue: Red proposes a feature, injects a subtle bug, and must repair it blind, while Blue reviews the pull request without knowing the bug and must pass all tests. A tournament of 11 model‑harness participants on 15 repositories across five languages showed GPT‑5.5/Codex CLI as the top solver and Claude Opus 4.8/mini‑swe‑agent as the top task author, with newer model versions producing harder tasks for older ones.
Agensh is a new multi‑agent harness that eliminates a central orchestrator by letting workers self‑organize through a continuous cooperation loop. The system uses a shared workspace, message interface, and shared context to coordinate tasks, verify results, and merge progress asynchronously. Experiments on ProgramBench and pandoc show that scaling from 1 to 1,024 agents improves test‑pass rates by up to 49% relative, demonstrating that agent count is a viable scaling dimension for complex tasks.
SWE‑Duel proposes a self‑scaling benchmark for coding agents, where tasks and oracles are generated at evaluation time to prevent data leakage and adapt difficulty to the agents’ capabilities. In the arena, two agents take turns modifying a real repository: the Red agent adds a feature, introduces a subtle bug, and the Blue agent must fix it without knowing the bug exists. A tournament of 11 model‑harness participants on 15 repositories in five languages showed GPT‑5.5 as the best solver and Claude Opus 4.8 as the best task author, with newer model versions producing harder tasks for older ones.
MARS (Multi-Agent Relay of Specialized LLMs) is a prompt-only framework that assigns specialized LLM agents—each focused on a particular algorithmic domain such as dynamic programming, graphs, or geometry—to collaboratively solve competitive programming problems. Retrieval-augmented generation selects a small team of relevant specialists for each problem, and the agents iteratively refine a C++17 solution through sandboxed testing, passing structured packets between them until a final infrastructure-fixer normalizes the code. On the CodeContests benchmark, MARS achieves a pass rate of 0.624 with Gemma 4, improving over direct prompting by 14.4 percentage points while reducing wall‑clock cost and token‑spend variance compared to CodeSIM.
arXiv:2610.08647v1 Announce Type: new Abstract: LLM-based agents solve complex multi-step tasks, but sequential execution incurs substantial latency. In principle, parallelizing work across multiple...
AgentRoom introduces a real‑time collaborative editing protocol that enables concurrent coding by multiple large language model agents within a CRDT‑backed shared workspace. By providing file‑level claim, status, and broadcast tools, it allows agents to coordinate directly rather than relying on serial phase handoffs or independent sampling. Experiments with five frontier coding‑CLI models show that AgentRoom reduces task abandonment and run‑to‑run variation compared to solo or parallel‑merge approaches, highlighting the importance of coordination over mere parallelism.
OpenCollab is a multi‑agent coding framework that unifies organization design, enforces experimental control on a shared runtime, and tracks execution via fine‑grained event streams. It introduces the metric Adherence to measure whether the declared organization is actually realized, showing that small configuration changes can shift adherence from 47.2% to 97.2%. Experiments demonstrate that a two‑coder workflow built on OpenCollab achieves new state‑of‑the‑art performance against mainstream harnesses while using the fewest tokens, and that a well‑designed organization can outperform strong existing harnesses.
arXiv:2608. 16801v1 Announce Type: new Abstract: We study how teams of AI coding agents coordinate while solving programming tasks.