{
  "schema_version": "1.1",
  "id": "s13:https://arxiv.org/abs/2609.02849v1",
  "slug": "2609-02849v1-1vwmx41",
  "url": "https://feed7.dev/p/2609-02849v1-1vwmx41",
  "title": "Post-Training Language Models for Gold-Medal Performance in Coding Competitions",
  "why_included": "A coding-specialized model paired post-training with an iterative generate-evaluate-refine loop to exceed the top IOI 2026 human score. The reusable idea is feedback-driven test-time search.",
  "summary": "The pipeline curated **22,000 problems** for SFT and RL, then used **GenCorrect** to generate, evaluate, and refine diverse solutions. Nano-CC rose from 130 to 468 points on IOI 2025; Ultra-CC reached 502.",
  "practical_implication": "For difficult coding tasks, allocate inference budget to multiple candidates, automated evaluation, and repair rather than relying on one completion. Specialized training data also mattered substantially before test-time search.",
  "agent_context": "The pipeline curated **22,000 problems** for SFT and RL, then used **GenCorrect** to generate, evaluate, and refine diverse solutions. Nano-CC rose from 130 to 468 points on IOI 2025; Ultra-CC reached 502.\n\nFor difficult coding tasks, allocate inference budget to multiple candidates, automated evaluation, and repair rather than relying on one completion. Specialized training data also mattered substantially before test-time search.\n\nThe prospective IOI 2026 system scored **535.4/600**, above the top human’s 498.27 under the same stated constraints. Competition performance does not establish comparable reliability on repository-scale engineering.",
  "source": {
    "name": "arXiv",
    "url": "https://arxiv.org/abs/2609.02849v1",
    "published_at": "2026-09-02T17:33:37.000Z"
  },
  "source_class": "blog_post",
  "content_type": "Paper",
  "layer": "model",
  "domains": [
    "coding"
  ],
  "topics": [
    "reasoning",
    "coding-agents",
    "model-selection"
  ],
  "verification": {
    "status": "needs_review",
    "label": "Needs Review",
    "method": "unverified",
    "verified_at": null
  },
  "uncertainty": [
    "The prospective IOI 2026 system scored **535.4/600**, above the top human’s 498.27 under the same stated constraints. Competition performance does not establish comparable reliability on repository-scale engineering."
  ],
  "connected_context": {
    "meaning": "This provides strong competition-specific evidence that specialized post-training and generate-evaluate-repair search can outperform a single completion. It shifts coding-model assessment toward the combined training-and-inference system, while explicitly limiting the result: exceptional algorithmic contest scores do not establish repository-scale engineering reliability.",
    "corpus_size": 669,
    "generated_at": "2026-09-03T10:01:29.442Z",
    "connections": [
      {
        "title": "The Base Model Is Dead — Varun Singh, Arcee AI",
        "source_name": "AI Engineer",
        "source_url": "https://www.youtube.com/watch?v=xbPriQWXtWM",
        "feed7_url": "https://feed7.dev/p/the-base-model-is-dead-varun-singh-arcee-ai-02hts76",
        "reason": "The curated coding data supports the claim that task-relevant capability is shaped upstream by data composition, not only by later routing or model branding."
      },
      {
        "title": "Introducing Grok 4.5",
        "source_name": "Cursor",
        "source_url": "https://cursor.com/blog/grok-4-5",
        "feed7_url": "https://feed7.dev/p/grok-4-5-1n0zgxx",
        "reason": "Grok 4.5’s benchmark exclusion highlights the evaluation-integrity checks needed before treating a striking coding score as comparative model-selection evidence."
      },
      {
        "title": "GLM 5.3 now available on AI Gateway",
        "source_name": "Vercel",
        "source_url": "https://vercel.com/changelog/glm-5-3-now-available-on-ai-gateway",
        "feed7_url": "https://feed7.dev/p/glm-5-3-now-available-on-ai-gateway-0s7o9zv",
        "reason": "GLM 5.3’s repository-scale positioning marks a different workload from programming contests, reinforcing the sourceBrief’s warning against transferring IOI performance directly to engineering agents."
      }
    ]
  },
  "lifecycle": "Current",
  "published_at": "2026-09-02T17:33:37.000Z",
  "modified_at": "2026-09-02T17:33:37.000Z",
  "supersedes": [],
  "expires_at": null,
  "formats": {
    "html": "https://feed7.dev/p/2609-02849v1-1vwmx41",
    "json": "https://feed7.dev/p/2609-02849v1-1vwmx41.json",
    "markdown": "https://feed7.dev/p/2609-02849v1-1vwmx41.md"
  }
}