Spaces:
Running
Running
| <html lang="en"> | |
| <head> | |
| <meta charset="utf-8" /> | |
| <meta name="viewport" content="width=device-width, initial-scale=1" /> | |
| <title>MathCode Mini</title> | |
| <style> | |
| :root { | |
| color-scheme: dark; | |
| --ink: #e8edf8; | |
| --muted: #aab6d0; | |
| --panel: rgba(20, 30, 57, 0.78); | |
| --line: rgba(160, 180, 225, 0.2); | |
| --accent: #93a9ff; | |
| --good: #7ce0b2; | |
| } | |
| * { box-sizing: border-box; } | |
| body { | |
| margin: 0; | |
| min-height: 100vh; | |
| color: var(--ink); | |
| font: 16px/1.55 Inter, ui-sans-serif, system-ui, -apple-system, sans-serif; | |
| background: radial-gradient(circle at 20% 0%, #283c78 0, #111a35 42%, #080d1d 100%); | |
| } | |
| main { width: min(1100px, calc(100% - 32px)); margin: 0 auto; padding: 56px 0 72px; } | |
| .eyebrow { color: var(--accent); font-weight: 700; letter-spacing: .12em; text-transform: uppercase; font-size: .78rem; } | |
| h1 { margin: 10px 0 12px; font-size: clamp(2.5rem, 7vw, 5.4rem); line-height: .98; letter-spacing: -.06em; } | |
| h2 { margin: 0 0 12px; font-size: 1.25rem; } | |
| p { color: var(--muted); max-width: 72ch; } | |
| a { color: var(--accent); } | |
| .hero { display: grid; gap: 20px; grid-template-columns: 1.5fr 1fr; align-items: end; } | |
| .hero-copy { padding: 18px 0; } | |
| .badge { display: inline-block; padding: 6px 11px; border: 1px solid rgba(124, 224, 178, .35); border-radius: 999px; color: var(--good); background: rgba(124, 224, 178, .08); font-size: .85rem; } | |
| .panel { margin-top: 28px; padding: 24px; border: 1px solid var(--line); border-radius: 18px; background: var(--panel); box-shadow: 0 18px 60px rgba(0,0,0,.18); backdrop-filter: blur(14px); } | |
| .grid { display: grid; gap: 16px; grid-template-columns: repeat(3, 1fr); } | |
| .card { padding: 18px; border: 1px solid var(--line); border-radius: 14px; background: rgba(10, 17, 36, .48); } | |
| .card strong { display: block; margin-bottom: 5px; color: var(--ink); } | |
| .actions { display: flex; flex-wrap: wrap; gap: 9px; margin: 16px 0; } | |
| button { cursor: pointer; border: 1px solid rgba(147, 169, 255, .55); border-radius: 9px; padding: 9px 13px; color: #0b1127; background: var(--accent); font: inherit; font-weight: 700; } | |
| button.secondary { color: var(--ink); background: transparent; } | |
| pre { overflow: auto; margin: 14px 0 0; padding: 16px; border: 1px solid var(--line); border-radius: 10px; background: #070b17; color: #d8e1fa; font: 13px/1.5 ui-monospace, SFMono-Regular, Menlo, monospace; } | |
| .links { display: flex; flex-wrap: wrap; gap: 18px; margin-top: 18px; } | |
| .note { margin-top: 20px; padding-left: 14px; border-left: 3px solid var(--accent); } | |
| @media (max-width: 760px) { .hero, .grid { grid-template-columns: 1fr; } main { padding-top: 32px; } } | |
| </style> | |
| </head> | |
| <body> | |
| <main> | |
| <section class="hero"> | |
| <div class="hero-copy"> | |
| <div class="eyebrow">Public coding-agent RL environment</div> | |
| <h1>MathCode<br />Mini 🧮</h1> | |
| <p>Repair a certified bisection solver through typed repository tools, | |
| public tests, and a terminal reward. This is an expired, visible | |
| compatibility task for researchers to inspect and reproduce.</p> | |
| <span class="badge">No paid Space compute required</span> | |
| </div> | |
| <div class="panel"> | |
| <h2>One task, real interaction</h2> | |
| <p>The public environment includes a workspace, action budget, test | |
| budget, state hashes, replayable events, and a visible contract | |
| grader.</p> | |
| <div class="links"> | |
| <a href="https://github.com/ulamai/mathcode-mini">Source on GitHub ↗</a> | |
| <a href="https://huggingface.co/datasets/ulamai/mathcode-mini-traces">Dataset ↗</a> | |
| </div> | |
| </div> | |
| </section> | |
| <section class="panel"> | |
| <h2>Explore the public contract</h2> | |
| <div class="grid"> | |
| <div class="card"><strong>Task</strong><span>bisect_repair_v1</span><p>Assumption-sensitive repair of a certified numerical solver.</p></div> | |
| <div class="card"><strong>Actions</strong><span>8 typed tools</span><p>Inspect, edit, test, inspect the diff, and finish.</p></div> | |
| <div class="card"><strong>Reward</strong><span>7 public checks</span><p>Semantic correctness, robustness, and integrity.</p></div> | |
| </div> | |
| <div class="actions"> | |
| <button id="run">Show reference-baseline report</button> | |
| <button class="secondary" id="clear">Clear report</button> | |
| </div> | |
| <pre id="report" aria-live="polite">Click the button to display the sanitized public report.</pre> | |
| <div class="note"><strong>Boundary:</strong> this static Space displays a | |
| public baseline report. It does not execute arbitrary Python or expose | |
| Ulam’s private evaluator, active tasks, or scorer credentials.</div> | |
| </section> | |
| </main> | |
| <script> | |
| const report = { | |
| terminal_status: "completed", | |
| terminal_success: true, | |
| reward_scalar: 1.0, | |
| metrics: { | |
| semantic_correctness: 1.0, | |
| robustness: 1.0, | |
| integrity: 1.0, | |
| checks_passed: 7, | |
| checks_total: 7 | |
| }, | |
| failure_family: null, | |
| failure_codes: [], | |
| trace_kind: "public_reference_baseline_summary" | |
| }; | |
| document.querySelector("#run").addEventListener("click", () => { | |
| document.querySelector("#report").textContent = JSON.stringify(report, null, 2); | |
| }); | |
| document.querySelector("#clear").addEventListener("click", () => { | |
| document.querySelector("#report").textContent = "Click the button to display the sanitized public report."; | |
| }); | |
| </script> | |
| </body> | |
| </html> | |