Decentralized Multi-Agent Systems with Shared Context

arXiv:2606.10662 · cs.MA, cs.AI · Submitted 2026-06-09 · Read on arXiv

cs.MA, cs.AI

Submitted: 2026-06-09

Updated: 2026-10-04

Code: https://github.com/datacurve-ai/deep-swe

Project page: https://yuzhenmao.github.io/DeLM

Terminology

• Multi-Agent System — 2× in this paper · explained in 6 episodes
• Accuracy — 1× in this paper · explained in 1 episode
• Test-time Scaling — 1× in this paper · explained in 2 episodes

Sources

• Oolong: Evaluating Long Context Reasoning and Aggregation Capabilities
• Large Language Monkeys: Scaling Inference Compute with Repeated Sampling
• Training Verifiers to Solve Math Word Problems
• DeepSeek-V4.1-Flash: Pushing the Limits of KV Cache Compression
• Improving Factuality and Reasoning in Language Models through Multiagent Debate
• AgentSwing: Adaptive Parallel Context Management Routing for Long-Horizon Web Agents
• Exploring Advanced LLM Multi-Agent Systems Based on Blackboard Architecture
• Kimi K2.5: Visual Agentic Intelligence
• A Human-Inspired Reading Agent with Gist Memory of Very Long Contexts
• Combee: Scaling Prompt Learning for Self-Improving Language Model Agents
• Encouraging Divergent Thinking in Large Language Models through Multi-Agent Debate
• Squeeze Evolve: Unified Multi-Model Orchestration for Verifier-Free Evolution
• AOrchestra: Automating Sub-Agent Creation for Agentic Orchestration
• LLM-Based Multi-Agent Blackboard System for Information Discovery in Data Science
• ToolOrchestra: Elevating Intelligence via Efficient Model and Tool Orchestration
• Self-Consistency Improves Chain of Thought Reasoning in Language Models
• SWE-agent: Agent-Computer Interfaces Enable Automated Software Engineering
• ProgramBench: Can Language Models Rebuild Programs From Scratch?
• Recursive Multi-Agent Systems
• Recursive Language Models

Related papers

• Highway Congestion Reduction through Reinforcement Learning Based Eulerian Headway Control
• ORACLE-SWE: Quantifying the Contribution of Critical Information Signals on SWE Agents
• You Only Align Once: Propagating Cooperative Behaviors in Multi-Agent Systems through Seed Agents
• Deny Without Disabling: Authorization-Paired Evaluation and Control for Multi-Agent Systems
• MA-SAPO: Multi-Agent Reasoning for Score-Aware Prompt Optimization
• PeroMAS: A Multi-agent System of Perovskite Material Discovery