AGENTIC-SYSTEMSJuly 7, 20261 min read

SwarmResearch: Orchestrating Coding Agents for Open-Ended Discovery

Researchers introduce SwarmResearch, a novel orchestrator-subagent framework that prevents coding agents from getting stuck in local optima by utilizing a central Shepherd Agent and multiple Search Agents operating on separate git branches.

Computer Science > Artificial Intelligence

Title:SwarmResearch: Orchestrating Coding Agents for Open-Ended Discovery

View PDF HTML (experimental)Abstract:Long-running coding agents such as autoresearch can persistently discover optimizations for open-ended problems. However, they tend to converge onto a single high-level approach, then proceed with low-level edits while missing other superior approaches to the problem. We hypothesize two harness-level design choices contribute to this behavior: accumulating context in a single long-running agent and only exposing a single program state to edit. We introduce SwarmResearch, an orchestrator-subagent harness in which a Shepherd Agent uses global context to steer a population of Search Agents, each operating with local context in their respective git branch. On open-ended optimization tasks, SwarmResearch discovers better or comparable solutions to state-of-the-art LLM-guided evolution and multi-agent techniques on 13/15 tasks, driven by higher-level exploration. Compared with fixed scaling of serial and parallel agents, SwarmResearch's orchestrator-guided scaling discovers better-performing solutions by adapting parallelism at different search depths.

Bibliographic and Citation Tools

Code, Data and Media Associated with this Article

Demos

Recommenders and Search Tools

arXivLabs: experimental projects with community collaborators

arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.

Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.

Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.

Source: arXiv cs.AI Recent

Stay in the loop

Get the best AI & tech insights in your inbox — no spam, unsubscribe anytime.

More in this category

agentic-systems

Automated Data Readiness for Scientific AI

Researchers have introduced REDI, an open-source framework that automates the transformation of raw scientific datasets into AI-ready data. Tested on the Frontier supercomputer, REDI addresses the data preparation bottleneck in scientific AI domains.

agentic-systems

ASK in the Dark: Uncertainty-Gated LLM Assistance under Partial Observability

Researchers have introduced ASK+, a novel framework that significantly enhances small language model (SLM) assistance for reinforcement learning agents under partial observability. By providing trajectory-aware context and structured reasoning, ASK+ demonstrates that prompt design and selective gating can outperform sheer model scale.

agentic-systems

VERITAS: Towards a General-Purpose Replication Tool for Scientific Research

As AI accelerates scientific publishing, independent verification has become increasingly difficult. VERITAS introduces a domain-agnostic replication framework powered by coding agents to automate experiment execution, code debugging, and claim verification.

agentic-systems

Oyster-II: Reinforcement Learning for Constructive Safety Alignment in Large Language Models

Researchers have introduced Oyster-II, a reinforcement learning-based constructive safety alignment framework for LLMs. By addressing the limitations of conventional refusal-oriented strategies, Oyster-II achieves an optimal balance between safety and helpfulness, rivaling much larger models.

agentic-systems

Internal Pluralism and the Limits of Pairwise Comparisons

A new study reveals that traditional pairwise comparisons used to align AI with human preferences have critical limitations due to the complexity of human decision-making. By modeling 'internal pluralism', researchers suggest that allowing users to express indecision can significantly improve preference-learning efficiency.

agentic-systems

iFLYTEK-Embodied-Omni Technical Report

iFLYTEK has introduced iFLYTEK-Embodied-Omni, a unified multimodal foundation model that integrates vision, language, and action. By establishing a unique 'brain-cerebellum' collaboration, it addresses the limitations of cascaded pipelines in embodied AI.