The article introduces SEAGym, a benchmark that treats self‑evolving LLM agents as reinforcement‑learning processes, evaluates their harness updates across multiple dimensions, and reveals how batch size, training source diversity, and backend model affect performance, stability, and cost.
EverMind’s newly released Raven Agent, built on the EverOS memory OS, showcases a self‑evolving, proactive AI that internalizes user interactions, offers 100,000 vetted skills, and aims to transition digital agents from stateless tools to autonomous digital life, marking a shift in AI memory and industry dynamics.
AReaL 2.0 introduces an online reinforcement‑learning infrastructure that turns real‑world agent interactions into a learning loop, defining three pillars—trajectory data protocol, data proxy, and evolution control plane—to enable agents to not only execute tasks but continuously improve from their own experience.
The article introduces GDPevo, the first open‑source benchmark that quantifies self‑evolution in agents by generating 120 real‑world enterprise tasks, using rule‑hybrid question creation and deterministic scoring, and shows that self‑evolving agents improve accuracy by 17‑22% while reducing token consumption.
SkillOpt reframes LLM agent skills as trainable external state, applying a deep‑learning‑style optimizer to systematically improve skill documents, and demonstrates across six benchmarks, seven models, and three execution modes that this approach yields consistent, large gains and robust transferability.
The paper introduces the Autogenesis Protocol (AGP), a two‑layer resource‑governed framework that lets agents safely modify their own prompts, tools, memory and environment, and demonstrates its effectiveness with the AGS system achieving 93.33% GAIA validation accuracy and near‑full scores on C++ LeetCode problems.
The article presents AgentScope Builder, a cloud‑native platform that extends the Harness framework to enable multi‑tenant, distributed, self‑evolving AI agents, detailing its workspace‑driven architecture, CompositeFilesystem isolation, zero‑code development flow, and how it scales from a single‑machine prototype (MinQwenPaw) to an enterprise‑grade solution.
The article clarifies that LangGraph, OpenClaw, and Hermes are not interchangeable products but three separate layers of agent development—Orchestration, User‑Facing entry, and Self‑evolving runtime—each solving different problems, with distinct use‑cases, strengths, and trade‑offs.
SkillOS reframes skill management for LLM agents as a long‑horizon reinforcement‑learning problem, letting a trainable Skill Curator automatically insert, update, or delete markdown‑based skills, which the frozen Agent Executor then consumes, improving memory‑free performance and cross‑task transfer.
This survey systematically reviews self‑evolving agents, explains why autonomous agents are needed, proposes a unified taxonomy of three evolution paradigms, analyzes model‑centric, environment‑centric, and co‑evolution approaches, and outlines future challenges in designing adaptive environments.
The DACon conference in Shanghai gathered over 8,000 developers, managers and experts, delivering 50 talks that explored self‑evolving AI agents, data‑centric ontology, Agent‑Ready big‑data infrastructure, AI‑AR ecosystem evolution, and the emerging challenges of Agentic data governance.
EverOS launches a global beta of its next‑generation memory infrastructure that lets autonomous agents automatically extract experience, cluster it semantically, and evolve reusable skills, boosting OpenClaw task success rates by up to 234.8% while addressing context‑window limits, multimodal retrieval, and developer transparency.
An in‑depth analysis of the OpenClaw‑derived Moltbook AI agent network reveals a “Self‑Evolution Trilemma” where continuous self‑evolution, complete isolation, and perpetual safety cannot coexist, supported by information‑theoretic definitions, empirical observations of cognitive decay, alignment failures, communication collapse, and proposed thermodynamic mitigation strategies.
