sora 13274243a0 Bump vendored EvalScope and add K3-ready DPV4 configs.
Keep K3 suite selection and report-schema scoring in bash, merge K3/vision dataset_args into dpv4 yamls, and pin EvalScope at 735d920ee911 with local patches.

Co-authored-by: Cursor <cursoragent@cursor.com>
2026-09-02 07:30:48 +00:00

4.0 KiB
Raw Permalink Blame History

ARC-AGI-2

概述

ARC-AGI-2面向人工通用智能的抽象与推理语料库 2是一个旨在衡量 AI 系统能否仅凭少量示例就高效地即时掌握新技能的基准测试。它通过网格变换任务来评估模型的抽象推理和模式识别能力。

任务描述

  • 任务类型:抽象推理 / 模式识别
  • 输入:一系列输入-输出网格对(示例),后跟一个测试输入网格
  • 输出:根据推断出的变换规则生成的预测输出网格
  • 网格格式二维整数数组0-9尺寸可变最大 30x30

主要特点

  • 包含 1,000 个公开训练任务和 120 个公开评估任务
  • 每个任务提供 2-10 个示例输入/输出对
  • 模型必须从示例中推断出变换规则
  • 测试抽象推理能力,不依赖于已学习的知识
  • 要求输出像素级精确(必须完全匹配目标网格)

评估说明

  • 评分基于精确网格匹配(形状和所有数值必须完全一致)
  • 模型必须以 JSON 二维数组格式输出网格
  • 零样本评估(每个任务内提供示例)
  • 设计为人类可解但对 AI 具有挑战性

属性

属性
基准测试名称 arc_agi_2
数据集ID evalscope/arc-agi-2
论文 N/A
标签 Reasoning
指标 accuracy
默认样本数 0-shot
评估划分 test
聚合方式 mean_and_pass_hat_k

数据统计

指标
总样本数 120
提示词长度(平均) 8026.79 字符
提示词长度(最小/最大) 2437 / 25471 字符

样例示例

子集: default

{
  "input": [
    {
      "id": "7b97143f",
      "content": "You are an expert at abstract reasoning and pattern recognition. Given input-output grid pairs as examples, you must figure out the transformation rule and apply it to a new test input to produce the correct output grid."
    },
    {
      "id": "bdaa2b22",
      "content": "You are given a series of input-output grid pairs as examples. Each grid is a 2D array of integers (0-9). Study the pattern in the examples, then predict the output for the test input.\n\nExamples:\nExample 1:\nInput: [[0, 0, 0, 0, 0, 0, 0, 0, 0, ... [TRUNCATED 7482 chars] ... 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0], [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0]]\n\nProvide the output grid as a JSON 2D array. Only output the JSON array, nothing else."
    }
  ],
  "target": "[[8, 0, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 0, 8, 8, 8, 8], [8, 0, 8, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 8, 0, 0, 0, 0, 0], [8, 0, 8, 0, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 0, 8, 8, 8, 8, 8, 0], [ ... [TRUNCATED 1596 chars] ... ], [8, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 8, 0, 8, 0, 0, 0, 8, 0], [8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 0, 8, 0, 8, 8, 8, 0], [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 8, 0, 8, 0, 0, 0]]",
  "id": 0,
  "group_id": 0
}

提示模板

系统提示

You are an expert at abstract reasoning and pattern recognition. Given input-output grid pairs as examples, you must figure out the transformation rule and apply it to a new test input to produce the correct output grid.

提示模板

{question}

使用方法

使用 CLI

evalscope eval \
    --model YOUR_MODEL \
    --api-url OPENAI_API_COMPAT_URL \
    --api-key EMPTY_TOKEN \
    --datasets arc_agi_2 \
    --limit 10  # 正式评估时请删除此行

使用 Python

from evalscope import run_task
from evalscope.config import TaskConfig

task_cfg = TaskConfig(
    model='YOUR_MODEL',
    api_url='OPENAI_API_COMPAT_URL',
    api_key='EMPTY_TOKEN',
    datasets=['arc_agi_2'],
    limit=10,  # 正式评估时请删除此行
)

run_task(task_cfg=task_cfg)