# Morpheus 基准：首个持续学习 RL 环境，LLM 表现不佳

- 来源：François Chollet (@fchollet)
- 发布时间：2026-07-14 01:26
- AIHOT 分数：36
- AIHOT 链接：https://aihot.virxact.com/items/cmrji6rvx0367bi03ke6cuvev
- 原文链接：https://x.com/fchollet/status/2076719958189613307

## AI 摘要

François Chollet 转发介绍 Morpheus，一个全新的持续学习基准。与 Atari、OpenAI Gym 等传统 RL 基准不同，Morpheus 提供永不重置的持久化企业模拟环境，目标异步变化，决策后果不断累积。团队测试了前沿 LLM 在 Morpheus 动态商业环境中的表现，主要结论是：LLM 并非持续学习者。论文及平台网站已公开。

## 正文

Standard RL benchmarks are episodic and stationary， so they don't capture the the characteristics of real-world deployment. Morpheus is a new benchmark for continual learning that provides persistent simulation environments where the world never resets， objectives shift asynchronously， and decisions have compounding consequences. Nice work by Sam & team！

Check out the paper - https://openreview.net/pdf?id=31P1VAfLkJ
Platform website - https://morpheus.skyfall.ai/

### 引用推文

> Skyfall AI：Today we present Morpheus, a persistent enterprise simulation platform designed to make Continual Learning a reality. Morpheus is the world's first real world R...