Agent-World: Scaling Real-World Environment Synthesis for Evolving General Agent Intelligence

Hugging Face Daily Papers Papers
ai-agents reinforcement-learning self-evolving large-language-models environment-synthesis mcp benchmark

Summary

Agent-World introduces a self-evolving training framework for general agent intelligence that autonomously discovers real-world environments and tasks via the Model Context Protocol, enabling continuous learning. Agent-World-8B and 14B models outperform strong proprietary models across 23 challenging agent benchmarks.

Large language models are increasingly expected to serve as general-purpose agents that interact with external, stateful tool environments. The Model Context Protocol (MCP) and broader agent skills offer a unified interface for connecting agents with scalable real-world services, but training robust agents remains limited by the lack of realistic environments and principled mechanisms for life-long learning. In this paper, we present Agent-World, a self-evolving training arena for advancing general agent intelligence through scalable environments. Agent-World has two main components: (1) Agentic Environment-Task Discovery, which autonomously explores topic-aligned databases and executable tool ecosystems from thousands of real-world environment themes and synthesizes verifiable tasks with controllable difficulty; and (2) Continuous Self-Evolving Agent Training, which combines multi-environment reinforcement learning with a self-evolving agent arena that automatically identifies capability gaps through dynamic task synthesis and drives targeted learning, enabling the co-evolution of agent policies and environments. Across 23 challenging agent benchmarks, Agent-World-8B and 14B consistently outperforms strong proprietary models and environment scaling baselines. Further analyses reveal scaling trends in relation to environment diversity and self-evolution rounds, offering insights for building general agent intelligence.
Original Article
View Cached Full Text

Cached at: 04/21/26, 07:20 AM

Paper page - Agent-World: Scaling Real-World Environment Synthesis for Evolving General Agent Intelligence

Source: https://huggingface.co/papers/2604.18292 Published on Apr 20

·

Submitted byhttps://huggingface.co/dongguanting

KABIon Apr 21

#2 Paper of the day Authors:

,

,

,

,

,

,

,

,

,

,

,

,

,

,

,

,

,

,

Abstract

Agent-World introduces a self-evolving training framework that advances general agent intelligence through autonomous environment discovery and continuous learning across diverse real-world scenarios.

Large language models are increasingly expected to serve as general-purpose agents that interact with external, stateful tool environments. TheModel Context Protocol(MCP) and broaderagent skillsoffer a unified interface for connecting agents with scalable real-world services, but training robust agents remains limited by the lack of realistic environments and principled mechanisms for life-long learning. In this paper, we present Agent-World, aself-evolving training arenafor advancing general agent intelligence through scalable environments. Agent-World has two main components: (1)Agentic Environment-Task Discovery, which autonomously explores topic-aligned databases and executable tool ecosystems from thousands of real-world environment themes and synthesizes verifiable tasks with controllable difficulty; and (2)Continuous Self-Evolving Agent Training, which combinesmulti-environment reinforcement learningwith a self-evolving agent arena that automatically identifies capability gaps throughdynamic task synthesisand drives targeted learning, enabling theco-evolution of agent policies and environments. Across 23 challenging agent benchmarks, Agent-World-8B and 14B consistently outperforms strong proprietary models and environment scaling baselines. Further analyses reveal scaling trends in relation to environment diversity and self-evolution rounds, offering insights for building general agent intelligence.

View arXiv pageView PDFProject pageAdd to collection

Get this paper in your agent:

hf papers read 2604\.18292

Don’t have the latest CLI?curl \-LsSf https://hf\.co/cli/install\.sh \| bash

Models citing this paper0

No model linking this paper

Cite arxiv.org/abs/2604.18292 in a model README.md to link it from this page.

Datasets citing this paper0

No dataset linking this paper

Cite arxiv.org/abs/2604.18292 in a dataset README.md to link it from this page.

Spaces citing this paper0

No Space linking this paper

Cite arxiv.org/abs/2604.18292 in a Space README.md to link it from this page.

Collections including this paper0

No Collection including this paper

Add this paper to acollectionto link it from this page.

Similar Articles

Qwen-AgentWorld: Language World Models for General Agents

Hacker News Top

Qwen-AgentWorld introduces language world models for agentic environments, covering seven domains with long chain-of-thought reasoning. The work includes a new benchmark, AgentWorldBench, and shows that world modeling improves downstream agent performance.

Multi-Agent World Models (3 minute read)

TLDR AI

γ-World is a generative multi-agent world model that supports independently controllable, permutation-symmetric agents using Simplex Rotary Agent Encoding and Sparse Hub Attention, achieving real-time 24 FPS rollouts and zero-shot generalization from two to four players.