index.html

<!doctype html>
<html lang="en">
<head>
<meta charset="utf-8">
<meta name="viewport" content="width=device-width,initial-scale=1">
<link rel="icon" href="data:image/svg+xml,%3Csvg xmlns='http://www.w3.org/2000/svg' viewBox='0 0 32 32'%3E%3Crect width='32' height='32' rx='6' fill='%230c1219'/%3E%3Ctext x='5' y='23' font-family='sans-serif' font-size='21' fill='%2378dbc0'%3EiN%3C/text%3E%3C/svg%3E">
<meta name="color-scheme" content="light dark">
<meta name="description" content="InferenceNet: a benchmark for AI agents that reproduce empirical research. Explore the data, agent harnesses and published leaderboard.">
<title>Home — InferenceNet Challenge</title>
<link rel="stylesheet" href="./leaderboard.css">
<link rel="stylesheet" href="./portal.css">
<script type="module" src="./portal.js">
</script>
</head>
<body>
<a class="skip" href="#main">
<span data-en="Skip to content" data-zh="跳转到正文" >Skip to content</span>
</a>
<div class="page">
<nav class="topbar" aria-label="Main navigation">
<a class="brand" href="./index.html">InferenceNet <small>CHALLENGE</small>
</a>
<div class="nav-links">
<a href="./index.html" aria-current="page">
<span data-en="Home" data-zh="主页" >Home</span>
</a>
<a href="./data.html">
<span data-en="Data" data-zh="数据" >Data</span>
</a>
<a href="./leaderboard.html">
<span data-en="Leaderboard" data-zh="榜单" >Leaderboard</span>
</a>
<a href="./agent.html">
<span data-en="Agent / Harness" data-zh="Agent / Harness" >Agent / Harness</span>
</a>
<button id="language" type="button" aria-label="Switch to Chinese">中文</button>
<button id="theme" type="button" aria-pressed="false">
<span class="theme-icon" aria-hidden="true">
</span>
<span id="theme-label">Light</span>
</button>
</div>
</nav>
<main id="main">
<header class="portal-hero home-hero">
<div>
<p class="eyebrow">INFERENCENET CHALLENGE</p>
<h1 data-en="Can AI reproduce an empirical finding?" data-zh="AI 能复现真实的实证研究吗？" >Can AI reproduce an empirical finding?</h1>
<p data-en="Real research data. Executable analysis. Results you can inspect." data-zh="真实研究数据，可执行的分析过程，可核查的结果。" class="hero-line">Real research data. Executable analysis. Results you can inspect.</p>
<p data-en="InferenceNet evaluates how models and agent harnesses turn an econometric specification into a reproducible result." data-zh="InferenceNet 评测模型与 Agent Harness 能否将计量经济学任务转化为可复现的研究结果。" class="portal-lead">InferenceNet evaluates how models and agent harnesses turn an econometric specification into a reproducible result.</p>
<div class="actions">
<a href="./leaderboard.html" class="action primary">
<span data-en="Explore the leaderboard →" data-zh="查看榜单 →" >Explore the leaderboard →</span>
</a>
<a href="./data.html" class="action">
<span data-en="Meet the data →" data-zh="了解数据 →" >Meet the data →</span>
</a>
</div>
</div>
<aside class="task-figure" aria-label="A replication task">
<div class="figure-label">
<span data-en="A REPLICATION TASK" data-zh="一道复现任务" >A REPLICATION TASK</span>
<span>0011 / OLS</span>
</div>
<div class="figure-question">
<h2 data-en="Economic uncertainty &amp; divisive politics" data-zh="经济不确定性与政治分歧" >Economic uncertainty &amp; divisive politics</h2>
<p data-en="Spain · 1905–1945 · month-clustered errors" data-zh="西班牙 · 1905–1945 · 按月聚类标准误" >Spain · 1905–1945 · month-clustered errors</p>
</div>
<div class="figure-stage">
<span class="file-chip">data_np.dta</span>
<span aria-hidden="true">+</span>
<span data-en="Research specification" data-zh="研究要求" >Research specification</span>
</div>
<div class="figure-arrow" aria-hidden="true">↓</div>
<div class="figure-model">Model + Harness</div>
<div class="figure-arrow" aria-hidden="true">↓</div>
<div class="output-fields">
<span>coefficient</span>
<span>standard_error</span>
<span>p_value</span>
</div>
<a href="./agent.html#case" class="figure-link">
<span data-en="Follow the recorded case ↗" data-zh="查看真实执行案例 ↗" >Follow the recorded case ↗</span>
</a>
</aside>
</header>
<div class="scale-strip">
<div>
<strong>1,000</strong>
<span data-en="fixed evaluation tasks" data-zh="固定评测任务" >fixed evaluation tasks</span>
</div>
<div>
<strong>15</strong>
<span data-en="journals in this subset" data-zh="子集涵盖期刊" >journals in this subset</span>
</div>
<div>
<strong>286</strong>
<span data-en="distinct article titles" data-zh="不同论文标题" >distinct article titles</span>
</div>
<div>
<strong>5</strong>
<span data-en="estimation classes" data-zh="估计方法大类" >estimation classes</span>
</div>
</div>
<p data-en="Selected_1000 · Counts from the pinned task list, revision 59f9512." data-zh="Selected_1000 · 按固定任务清单统计，版本 59f9512。" class="source-note">Selected_1000 · Counts from the pinned task list, revision 59f9512.</p>
<section class="portal-section">
<div class="portal-heading">
<p data-en="01 / THE CHALLENGE" data-zh=" 01 / THE CHALLENGE" class="eyebrow">01 / THE CHALLENGE</p>
<h2 data-en="From a research question to a reproducible result." data-zh="从研究问题，到可复现的结果。" >From a research question to a reproducible result.</h2>
</div>
<ol class="workflow">
<li>
<span class="step-number">01</span>
<h3 data-en="Read the task" data-zh="读取任务" >Read the task</h3>
<p data-en="A research specification and its source data." data-zh="研究要求与对应的数据文件。" >A research specification and its source data.</p>
</li>
<li>
<span class="step-number">02</span>
<h3 data-en="Write &amp; run code" data-zh="编写并执行代码" >Write &amp; run code</h3>
<p data-en="Fit the requested econometric model." data-zh="按照要求估计计量模型。" >Fit the requested econometric model.</p>
</li>
<li>
<span class="step-number">03</span>
<h3 data-en="Inspect &amp; revise" data-zh="检查并修订" >Inspect &amp; revise</h3>
<p data-en="A harness can use execution feedback." data-zh="Harness 可根据执行反馈修订程序。" >A harness can use execution feedback.</p>
</li>
<li>
<span class="step-number">04</span>
<h3 data-en="Recover the finding" data-zh="复现研究结果" >Recover the finding</h3>
<p data-en="Return the coefficient, standard error and p-value." data-zh="输出系数、标准误和 p 值。" >Return the coefficient, standard error and p-value.</p>
</li>
</ol>
</section>
<section class="portal-section">
<div class="portal-heading">
<p data-en="02 / EXPLORE" data-zh=" 02 / EXPLORE" class="eyebrow">02 / EXPLORE</p>
<h2 data-en="One project. Three ways in." data-zh="从数据、结果与执行过程了解项目。" >One project. Three ways in.</h2>
</div>
<div class="entry-grid">
<a class="entry" href="./data.html">
<span class="eyebrow">01</span>
<h3 data-en="Explore the data" data-zh="了解数据" >Explore the data</h3>
<p data-en="Where the tasks come from, what they ask, and how to access them." data-zh="了解任务来源、研究要求与数据获取方式。" >Where the tasks come from, what they ask, and how to access them.</p>
<span class="entry-arrow" aria-hidden="true">↗</span>
</a>
<a class="entry" href="./leaderboard.html">
<span class="eyebrow">02</span>
<h3 data-en="Compare the results" data-zh="比较结果" >Compare the results</h3>
<p data-en="Single-pass models, DeepAgents and an independent DSH run." data-zh="单次生成、DeepAgents，以及独立的 DSH 实验。" >Single-pass models, DeepAgents and an independent DSH run.</p>
<span class="entry-arrow" aria-hidden="true">↗</span>
</a>
<a class="entry" href="./agent.html">
<span class="eyebrow">03</span>
<h3 data-en="Inside the agent loop" data-zh="走进 Agent 循环" >Inside the agent loop</h3>
<p data-en="Follow the code, tools and final artifact of a real replication task." data-zh="通过真实案例了解代码、工具与最终产物。" >Follow the code, tools and final artifact of a real replication task.</p>
<span class="entry-arrow" aria-hidden="true">↗</span>
</a>
</div>
</section>
<section class="portal-section">
<div class="portal-heading">
<p data-en="03 / RESULTS" data-zh=" 03 / RESULTS" class="eyebrow">03 / RESULTS</p>
<h2 data-en="A shared view of the evidence." data-zh="在同一份证据上比较。" >A shared view of the evidence.</h2>
</div>
<div class="release-strip">
<div>
<p class="eyebrow">
<span data-en="PUBLISHED RESEARCH SNAPSHOT" data-zh="已发布的研究快照" >PUBLISHED RESEARCH SNAPSHOT</span>
</p>
<div class="release-count">
<strong data-snapshot="groups">—</strong> <span data-en="experiment groups" data-zh="组实验" >experiment groups</span> <span class="muted">/</span> <strong data-snapshot="models">—</strong> <span data-en="model aliases" data-zh="个模型别名" >model aliases</span>
</div>
<p data-en="The homepage and leaderboard read the same published result file." data-zh="主页与榜单使用同一份已发布结果文件。" >The homepage and leaderboard read the same published result file.</p>
<p class="snapshot-date" data-snapshot="date" aria-live="polite">
</p>
</div>
<a href="./leaderboard.html" class="action">
<span data-en="View results →" data-zh="打开榜单 →" >View results →</span>
</a>
</div>
<p data-en="Research results use local scoring. Read the run protocols, scoring definitions and evidence status with every comparison." data-zh="研究结果采用本地评分；比较时请同时查看运行协议、指标定义与证据状态。" class="source-note">Research results use local scoring. Read the run protocols, scoring definitions and evidence status with every comparison.</p>
</section>
</main>
<footer>
<span>InferenceNet <span class="muted">/</span> <span data-en="AI for empirical research." data-zh="面向实证研究的 AI。" >AI for empirical research.</span>
</span>
<div>
<a href="https://huggingface.co/datasets/CamoAiLab/InferenceNet" class="" target="_blank" rel="noopener noreferrer">
<span data-en="Dataset ↗" data-zh="数据集 ↗" >Dataset ↗</span>
</a>
<a href="https://easonai-5589.github.io/inferencenet-challenge/" class="" target="_blank" rel="noopener noreferrer">
<span data-en="Project website ↗" data-zh="项目官网 ↗" >Project website ↗</span>
</a>
<a href="https://easonai-5589.github.io/inferencenet-challenge/teams.html" class="" target="_blank" rel="noopener noreferrer">
<span data-en="Team &amp; contact ↗" data-zh="团队与联系 ↗" >Team &amp; contact ↗</span>
</a>
</div>
</footer>
</div>
</body>
</html>