# ARC Prize 2024：测试时训练潜力巨大

- 来源：François Chollet (@fchollet)
- 发布时间：2026-08-13 16:41
- AIHOT 分数：54
- AIHOT 链接：https://aihot.virxact.com/items/cmsra3mff0qvoroz2ekcqkp5h
- 原文链接：https://x.com/fchollet/status/2087821894581973426

## AI 摘要

François Chollet 认为测试时训练（TTT）潜力巨大，并希望其走向主流。ARC Prize 2024 将 SOTA 从 33% 提升至 55.5%，创 2020 年以来最大年度增幅，TTT 是自 2019-2020 年上下文学习以来 LLM 泛化能力的最大跃升。冠军为 the ARChitects 团队。

## 正文

Test-time training was popularized during the ARC Prize 2024 competition, after being explored in particular by @MindsAI_Jack and team. To date, I believe ARC 1-2 are the only datasets where TTT strongly outperforms. It would be interesting if TTT started becoming more mainstream. I believe it has great potential.

### 引用推文

> François Chollet：Today we're announcing the winners of ARC Prize 2024. We're also publishing an extensive technical report on what we learned from the competition (link in the n...