它使用 Claude Opus 5,但不更新模型权重,也不针对单款游戏微调;主要做法是把对游戏规则的理解写入外部文档和可执行代码。
智能体通过“观察—反思—修订规则—编译代码—验证”的循环,逐步完善世界模型,再用它预演并规划行动。
这项结果针对公开测试集;操作次数较少不等于总推理成本更低,也不能据此断言系统已能应对各种真实环境。
How did Memento 3, developed by Huawei Noah’s Ark Lab, University College London and partners, clear all 25 games and 183 levels of the ARCAI-generated editorial hero image for How did Memento 3, developed by Huawei Noah’s Ark Lab, University College London and partners, clear all 25 games and 183 levels of the ARC.
AI 提示
Create a landscape editorial hero image for this Studio Global article: How did Memento 3, developed by Huawei Noah’s Ark Lab, University College London and partners, clear all 25 games and 183 levels of the ARC. Article summary: Memento 3’s reported gain came from changing the agent’s external memory and planning machinery—not the language model’s weights: it repeatedly wrote, tested and revised a rulebook and executable simulator, then used tha. Topic tags: general web, llm, ai, workflow, productivity. Style: premium digital editorial illustration, source-backed research mood, clean composition, high detail, modern web publication hero. Use reference image context only for broad subject, composition, and topical grounding; do not copy the exact image. Avoid: logos, brand marks, copyrighted characters, real person likenesses, fake screenshots, UI text, readable text, watermarks, charts
关键在于,这项提升并非来自重新训练底层语言模型。报告称,系统使用 Claude Opus 5,运行过程中没有更新模型权重,也没有针对具体游戏微调。101314 在 ARC Prize 的官方测试框架中,Claude Opus 5 的 RHAE 据报约为 40.7;不过,这一对照能说明整套智能体架构的表现差异,并不能单独证明每个组成部分分别带来了多少提升。1014
它使用 Claude Opus 5,但不更新模型权重,也不针对单款游戏微调;主要做法是把对游戏规则的理解写入外部文档和可执行代码。
智能体通过“观察—反思—修订规则—编译代码—验证”的循环,逐步完善世界模型,再用它预演并规划行动。
这项结果针对公开测试集;操作次数较少不等于总推理成本更低,也不能据此断言系统已能应对各种真实环境。
How did Memento 3, developed by Huawei Noah’s Ark Lab, University College London and partners, clear all 25 games and 183 levels of the ARCAI-generated editorial hero image for How did Memento 3, developed by Huawei Noah’s Ark Lab, University College London and partners, clear all 25 games and 183 levels of the ARC.
AI 提示
Create a landscape editorial hero image for this Studio Global article: How did Memento 3, developed by Huawei Noah’s Ark Lab, University College London and partners, clear all 25 games and 183 levels of the ARC. Article summary: Memento 3’s reported gain came from changing the agent’s external memory and planning machinery—not the language model’s weights: it repeatedly wrote, tested and revised a rulebook and executable simulator, then used tha. Topic tags: general web, llm, ai, workflow, productivity. Style: premium digital editorial illustration, source-backed research mood, clean composition, high detail, modern web publication hero. Use reference image context only for broad subject, composition, and topical grounding; do not copy the exact image. Avoid: logos, brand marks, copyrighted characters, real person likenesses, fake screenshots, UI text, readable text, watermarks, charts
关键在于,这项提升并非来自重新训练底层语言模型。报告称,系统使用 Claude Opus 5,运行过程中没有更新模型权重,也没有针对具体游戏微调。101314 在 ARC Prize 的官方测试框架中,Claude Opus 5 的 RHAE 据报约为 40.7;不过,这一对照能说明整套智能体架构的表现差异,并不能单独证明每个组成部分分别带来了多少提升。1014