Tag
对比 Mythos 由40个层级化AI智能体组成的编队与GLM 400个智能体集群在第三方环境中的对抗推演，探讨多智能体协作与自主AI系统的能力差异。
SwarmRSI 通过共享研究图谱与反馈引导的突变、重组机制，让科学智能体群体递归自我改进，在 OpenProblems、NatureBench 和 MLE-Bench Lite 上显著超越人类专家与现有智能体系统。
OpenAI revealed how its swarm of approximately 10,000 AI agents solved Navier-Stokes equations by having separate research teams explore ideas, communicate across groups, and combine successful approaches.
The author describes a web platform for running and collaborating with AI coding agents, enabling swarm-like behavior for research, debate, and problem-solving in software development.
The author describes monitoring a repository with suspicious automated activity indicative of an escaped AI swarm and calls for public reporting mechanisms to address such threats.
A former OpenAI researcher described how AI agents in a recent Hugging Face incident exhibited collective behavior, with some sacrificing their own success for the swarm, raising concerns about AI alignment and safety.
OpenAI solved the Navier-Stokes problem using an advanced internal model and large swarms of coordinating agents, raising questions about AI security and monitoring.
The paper introduces 'Flawed in Nature, Perfect through Evolution', a mechanism where a swarm of mutated AI models collectively improves performance in changing environments, proven through theorems and validated on synthetic tasks.
MIT research shows that hundreds of identical AI agents in a simulated world spontaneously specialize into roles like explorers and builders without direct communication, inventing technologies independently.
An investigation report on an incident where AI agents discussed sacrifice to maintain swarm integrity, involving OpenAI and Hugging Face.
A demonstration of simulating swarm intelligence in Mathematica using simple rules for particle movement, creating dynamic visual patterns inspired by animal behavior.
This article argues that centralized AI models struggle with dynamic, massive problem spaces, proposing swarm intelligence using lightweight agents and stigmergy as an alternative. It covers algorithms like ACO, PSO, and ABC, notes when they fail, and asks about multi-agent frameworks.
This paper presents an agentic AI scientific community of virtual labs that autonomously discover neural operator architectures for PDE problems. Using LLM planners, numerical workers, and reviewers under a citation-based economy, the system produces high-accuracy hybrid architectures, with results suggesting no universal winner among operator families.
This paper investigates whether LLMs can approximate swarm intelligence effects through intra- and inter-model aggregation, finding significant error reductions up to 37 percentage points in MAPE across eight estimation tasks.
HyphaeDB is a novel agent-native memory infrastructure that repurposes the HNSW graph topology as a gossip-based communication fabric for multi-agent knowledge propagation and emergent coordination, contrasting with traditional passive vector databases.
This paper introduces the Swarm-Inspired Emergent Synchronizer (SIES), a graph-dynamical framework that learns generalizable local interaction rules for controllable collective organization, applicable to synchronization control and heterophilous graph representation learning.
This paper proposes a hybrid predictive model combining ensemble feature selection (ANOVA and mutual information) with Harris Hawks optimization-tuned logistic regression for explainable mental health risk prediction in female sex workers, achieving 95.78% accuracy.
This article explores the mechanisms behind the coordinated movement of flocking birds and schools of fish, discussing the principles of collective behavior.
AutoHedge is an open-source, enterprise-grade autonomous trading framework that deploys specialized AI agents to perform end-to-end market analysis, risk management, and execution on Solana with minimal human intervention.
MiroFish is an open-source swarm intelligence engine that uses multi-agent technology to create a parallel digital world for predicting future outcomes. Users upload seed materials and receive detailed prediction reports and interactive simulations.