liqiang/ai-agent-book
1
0
Fork 0
Code Issues Pull Requests Actions 1 Packages Projects Releases Wiki Activity
Files
main
ai-agent-book/book/images/fig8-reward-paradigms.svg
T
liqiang b119135836
Build latest book artifacts / build (push) Canceled after 0s
Details
dependency resolution / resolve (3.11) (push) Canceled after 0s
Details
dependency resolution / resolve (3.13) (push) Canceled after 0s
Details
deploy-pages / build (push) Canceled after 0s
Details
deploy-pages / deploy (push) Canceled after 0s
Details
i18n consistency check / check (push) Canceled after 0s
Details
provider adoption tests / test (chapter2/context-compression) (push) Canceled after 0s
Details
provider adoption tests / test (chapter2/prompt-injection) (push) Canceled after 0s
Details
provider adoption tests / test (chapter2/system-hint) (push) Canceled after 0s
Details
provider adoption tests / test (chapter3/log-sanitization) (push) Canceled after 0s
Details
web-search-agent tests / test (push) Canceled after 0s
Details
web-search-agent tests / agentbook (push) Canceled after 0s
Details
ai-agent-book 精选快照（<2MB 代码与文档，来自 github.com/bojieli/ai-agent-book）
2026-08-20 13:12:50 +00:00

54 lines
9.7 KiB
XML
Raw Permalink Blame History

This file contains ambiguous Unicode characters
This file contains Unicode characters that might be confused with other characters. If you think that this is intentional, you can safely ignore this warning. Use the Escape button to reveal them.
<svg xmlns="http://www.w3.org/2000/svg" viewBox="0 40 780 500" width="780" height="500" style="background:#ffffff">
<defs><marker id="ah" markerWidth="12" markerHeight="8" refX="12" refY="4" orient="auto"><polygon points="0 0, 12 4, 0 8" fill="#333333"/></marker><marker id="ah-light" markerWidth="12" markerHeight="8" refX="12" refY="4" orient="auto"><polygon points="0 0, 12 4, 0 8" fill="#999999"/></marker></defs>
<rect x="50" y="60" width="380" height="78" rx="4" fill="#f0f0f0" stroke="#333333" stroke-width="2"/>
<text x="60" y="78" font-family="Arial, 'Helvetica Neue', Helvetica, 'PingFang SC', 'Microsoft YaHei', sans-serif" font-size="16" fill="#333333" text-anchor="start" dominant-baseline="central" font-weight="bold">文本思考₁</text>
<text x="65" y="98" font-family="Arial, 'Helvetica Neue', Helvetica, 'PingFang SC', 'Microsoft YaHei', sans-serif" font-size="14" fill="#666666" text-anchor="start" dominant-baseline="central" font-weight="normal">&quot;需要求解方程 x²+2x-8=0</text>
<text x="65" y="114" font-family="Arial, 'Helvetica Neue', Helvetica, 'PingFang SC', 'Microsoft YaHei', sans-serif" font-size="14" fill="#666666" text-anchor="start" dominant-baseline="central" font-weight="normal">先用 sympy 求解...&quot;</text>
<line x1="240" y1="139" x2="240" y2="145" stroke="#999999" stroke-width="2" marker-end="url(#ah-light)"/>
<rect x="50" y="146" width="380" height="78" rx="4" fill="#f5f5f5" stroke="#333333" stroke-width="2"/>
<text x="60" y="164" font-family="Arial, 'Helvetica Neue', Helvetica, 'PingFang SC', 'Microsoft YaHei', sans-serif" font-size="16" fill="#333333" text-anchor="start" dominant-baseline="central" font-weight="bold">代码生成₁</text>
<text x="65" y="178" font-family="'Courier New', Courier, monospace" font-size="11" fill="#333333" text-anchor="start" dominant-baseline="central">&lt;code&gt;</text>
<text x="65" y="188" font-family="'Courier New', Courier, monospace" font-size="11" fill="#333333" text-anchor="start" dominant-baseline="central">from sympy import *</text>
<text x="65" y="198" font-family="'Courier New', Courier, monospace" font-size="11" fill="#333333" text-anchor="start" dominant-baseline="central">x=symbols(&quot;x&quot;)</text>
<text x="65" y="208" font-family="'Courier New', Courier, monospace" font-size="11" fill="#333333" text-anchor="start" dominant-baseline="central">solve(x**2+2*x-8,x)</text>
<text x="65" y="218" font-family="'Courier New', Courier, monospace" font-size="11" fill="#333333" text-anchor="start" dominant-baseline="central">&lt;/code&gt;</text>
<line x1="240" y1="225" x2="240" y2="231" stroke="#999999" stroke-width="2" marker-end="url(#ah-light)"/>
<rect x="50" y="232" width="380" height="78" rx="4" fill="#d0d0d0" stroke="#333333" stroke-width="2"/>
<text x="60" y="250" font-family="Arial, 'Helvetica Neue', Helvetica, 'PingFang SC', 'Microsoft YaHei', sans-serif" font-size="16" fill="#333333" text-anchor="start" dominant-baseline="central" font-weight="bold">沙箱反馈₁</text>
<text x="65" y="270" font-family="'Courier New', Courier, monospace" font-size="12" fill="#333333" text-anchor="start" dominant-baseline="central">&lt;interpreter&gt;</text>
<text x="65" y="286" font-family="'Courier New', Courier, monospace" font-size="12" fill="#333333" text-anchor="start" dominant-baseline="central">[-4, 2]</text>
<text x="65" y="302" font-family="'Courier New', Courier, monospace" font-size="12" fill="#333333" text-anchor="start" dominant-baseline="central">&lt;/interpreter&gt;</text>
<line x1="240" y1="311" x2="240" y2="317" stroke="#999999" stroke-width="2" marker-end="url(#ah-light)"/>
<rect x="50" y="318" width="380" height="78" rx="4" fill="#f0f0f0" stroke="#333333" stroke-width="2"/>
<text x="60" y="336" font-family="Arial, 'Helvetica Neue', Helvetica, 'PingFang SC', 'Microsoft YaHei', sans-serif" font-size="16" fill="#333333" text-anchor="start" dominant-baseline="central" font-weight="bold">文本思考₂</text>
<text x="65" y="356" font-family="Arial, 'Helvetica Neue', Helvetica, 'PingFang SC', 'Microsoft YaHei', sans-serif" font-size="14" fill="#666666" text-anchor="start" dominant-baseline="central" font-weight="normal">&quot;解为 x=-4 或 x=2</text>
<text x="65" y="372" font-family="Arial, 'Helvetica Neue', Helvetica, 'PingFang SC', 'Microsoft YaHei', sans-serif" font-size="14" fill="#666666" text-anchor="start" dominant-baseline="central" font-weight="normal">验证：2²+2×2-8=0 ✓&quot;</text>
<line x1="240" y1="397" x2="240" y2="403" stroke="#999999" stroke-width="2" marker-end="url(#ah-light)"/>
<rect x="50" y="404" width="380" height="78" rx="4" fill="#d0d0d0" stroke="#333333" stroke-width="2"/>
<text x="60" y="422" font-family="Arial, 'Helvetica Neue', Helvetica, 'PingFang SC', 'Microsoft YaHei', sans-serif" font-size="16" fill="#333333" text-anchor="start" dominant-baseline="central" font-weight="bold">最终答案</text>
<text x="65" y="442" font-family="Arial, 'Helvetica Neue', Helvetica, 'PingFang SC', 'Microsoft YaHei', sans-serif" font-size="14" fill="#666666" text-anchor="start" dominant-baseline="central" font-weight="normal">\boxed{-4, 2}</text>
<rect x="470" y="60" width="280" height="130" rx="6" fill="#f0f0f0" stroke="#333333" stroke-width="2"/>
<text x="610" y="82" font-family="Arial, 'Helvetica Neue', Helvetica, 'PingFang SC', 'Microsoft YaHei', sans-serif" font-size="20" fill="#333333" text-anchor="middle" dominant-baseline="central" font-weight="bold">SandboxFusion</text>
<text x="610" y="105" font-family="Arial, 'Helvetica Neue', Helvetica, 'PingFang SC', 'Microsoft YaHei', sans-serif" font-size="16" fill="#666666" text-anchor="middle" dominant-baseline="central" font-weight="normal">128 并行 worker</text>
<text x="610" y="125" font-family="Arial, 'Helvetica Neue', Helvetica, 'PingFang SC', 'Microsoft YaHei', sans-serif" font-size="16" fill="#666666" text-anchor="middle" dominant-baseline="central" font-weight="normal">Python 沙箱执行</text>
<text x="610" y="148" font-family="Arial, 'Helvetica Neue', Helvetica, 'PingFang SC', 'Microsoft YaHei', sans-serif" font-size="14" fill="#666666" text-anchor="middle" dominant-baseline="central" font-weight="normal">sympy/numpy/scipy</text>
<text x="610" y="168" font-family="Arial, 'Helvetica Neue', Helvetica, 'PingFang SC', 'Microsoft YaHei', sans-serif" font-size="14" fill="#666666" text-anchor="middle" dominant-baseline="central" font-weight="normal">超时控制 + 内存限制</text>
<path d="M 432,185 Q 432.5041728782917,142.81000344028462 468,120" fill="none" stroke="#333333" stroke-width="2" marker-end="url(#ah)"/>
<rect x="422" y="126" width="40" height="24" rx="3" fill="white"/>
<text x="441.25208643914584" y="137.6550017201423" font-family="Arial, 'Helvetica Neue', Helvetica, 'PingFang SC', 'Microsoft YaHei', sans-serif" font-size="16" fill="#666666" text-anchor="middle" dominant-baseline="central" font-weight="normal">代码</text>
<path d="M 468,150 Q 469.1695571333223,216.2033393123934 432,271" fill="none" stroke="#333333" stroke-width="2" marker-end="url(#ah)"/>
<rect x="440" y="192" width="40" height="24" rx="3" fill="white"/>
<text x="459.58477856666116" y="203.3516696561967" font-family="Arial, 'Helvetica Neue', Helvetica, 'PingFang SC', 'Microsoft YaHei', sans-serif" font-size="16" fill="#666666" text-anchor="middle" dominant-baseline="central" font-weight="normal">结果</text>
<rect x="470" y="220" width="280" height="260" rx="4" fill="#f5f5f5" stroke="#333333" stroke-width="2"/>
<text x="610" y="242" font-family="Arial, 'Helvetica Neue', Helvetica, 'PingFang SC', 'Microsoft YaHei', sans-serif" font-size="20" fill="#333333" text-anchor="middle" dominant-baseline="central" font-weight="bold">PPO 训练循环</text>
<text x="480" y="265" font-family="Arial, 'Helvetica Neue', Helvetica, 'PingFang SC', 'Microsoft YaHei', sans-serif" font-size="14" fill="#333333" text-anchor="start" dominant-baseline="central" font-weight="bold">基础模型: Qwen2.5-32B</text>
<text x="480" y="283" font-family="Arial, 'Helvetica Neue', Helvetica, 'PingFang SC', 'Microsoft YaHei', sans-serif" font-size="14" fill="#666666" text-anchor="start" dominant-baseline="central" font-weight="normal">SFT 冷启动: ~1 小时</text>
<text x="480" y="301" font-family="Arial, 'Helvetica Neue', Helvetica, 'PingFang SC', 'Microsoft YaHei', sans-serif" font-size="14" fill="#666666" text-anchor="start" dominant-baseline="central" font-weight="normal">RL 训练: ~9 天 / 400 步</text>
<text x="480" y="337" font-family="Arial, 'Helvetica Neue', Helvetica, 'PingFang SC', 'Microsoft YaHei', sans-serif" font-size="14" fill="#666666" text-anchor="start" dominant-baseline="central" font-weight="normal">每步: 32题 × 16候选 = 512</text>
<text x="480" y="355" font-family="Arial, 'Helvetica Neue', Helvetica, 'PingFang SC', 'Microsoft YaHei', sans-serif" font-size="14" fill="#666666" text-anchor="start" dominant-baseline="central" font-weight="normal">平均 7-9 轮交互/响应</text>
<text x="480" y="391" font-family="Arial, 'Helvetica Neue', Helvetica, 'PingFang SC', 'Microsoft YaHei', sans-serif" font-size="14" fill="#333333" text-anchor="start" dominant-baseline="central" font-weight="bold">AIME 2024 准确率:</text>
<text x="480" y="409" font-family="Arial, 'Helvetica Neue', Helvetica, 'PingFang SC', 'Microsoft YaHei', sans-serif" font-size="14" fill="#666666" text-anchor="start" dominant-baseline="central" font-weight="normal"> SFT 基线: 25%</text>
<text x="480" y="427" font-family="Arial, 'Helvetica Neue', Helvetica, 'PingFang SC', 'Microsoft YaHei', sans-serif" font-size="14" fill="#666666" text-anchor="start" dominant-baseline="central" font-weight="normal"> 110 步后: 52%</text>
<text x="480" y="445" font-family="Arial, 'Helvetica Neue', Helvetica, 'PingFang SC', 'Microsoft YaHei', sans-serif" font-size="14" fill="#666666" text-anchor="start" dominant-baseline="central" font-weight="normal"> 400 步后: 67%</text>
</svg>
Reference in New Issue View Git Blame Copy Permalink