{
  "schema_version": "1.0",
  "id": "archive:https://arxiv.org/abs/2607.02512v1",
  "slug": "2607-02512v1-1dr5458",
  "url": "https://feed7.dev/p/2607-02512v1-1dr5458",
  "title": "Program-as-Weights: A Programming Paradigm for Fuzzy Functions",
  "why_included": "Program-as-Weights compiles natural-language fuzzy functions (JSON repair, log filtering) into adapters for a frozen 0.6B interpreter — matching Qwen3-32B prompting at ~1/50th the memory, 30 tok/s on an M3.",
  "summary": "**The gist** A **4B compiler model** trained on **FuzzyBench (10M examples)** turns a natural-language spec into a parameter-efficient adapter that runs on a frozen **0.6B Qwen3** interpreter. The compiled function matches direct prompting of **Qwen3-32B** at about **1/50th the inference memory**, hitting **30 tokens/s on a MacBook M3**.",
  "practical_implication": "**Why it matters** The fuzzy glue you currently route to an LLM API — repairing malformed JSON, flagging important log lines, ranking by intent — could become a compile-once, run-locally artifact: reproducible, cheap, offline. It reframes big models as **tool builders** invoked once per function definition rather than once per call.",
  "agent_context": "**The gist** A **4B compiler model** trained on **FuzzyBench (10M examples)** turns a natural-language spec into a parameter-efficient adapter that runs on a frozen **0.6B Qwen3** interpreter. The compiled function matches direct prompting of **Qwen3-32B** at about **1/50th the inference memory**, hitting **30 tokens/s on a MacBook M3**.\n\n**Why it matters** The fuzzy glue you currently route to an LLM API — repairing malformed JSON, flagging important log lines, ranking by intent — could become a compile-once, run-locally artifact: reproducible, cheap, offline. It reframes big models as **tool builders** invoked once per function definition rather than once per call.\n\n**Watch out** Results come from tasks resembling the **FuzzyBench** training distribution; how compilation holds for messier or novel specs, and how you validate a compiled function's behavior before trusting it, is untested here.",
  "source": {
    "name": "arXiv",
    "url": "https://arxiv.org/abs/2607.02512v1",
    "published_at": null
  },
  "source_class": "blog_post",
  "content_type": "Paper",
  "layer": "model",
  "domains": [
    "coding",
    "research"
  ],
  "topics": [
    "open-models",
    "model-selection"
  ],
  "verification": {
    "status": "needs_review",
    "label": "Needs Review",
    "method": "unverified",
    "verified_at": null
  },
  "uncertainty": [
    "Results come from tasks resembling the **FuzzyBench** training distribution; how compilation holds for messier or novel specs, and how you validate a compiled function's behavior before trusting it, is untested here."
  ],
  "lifecycle": "Current",
  "published_at": null,
  "modified_at": null,
  "supersedes": [],
  "expires_at": null,
  "formats": {
    "html": "https://feed7.dev/p/2607-02512v1-1dr5458",
    "json": "https://feed7.dev/p/2607-02512v1-1dr5458.json",
    "markdown": "https://feed7.dev/p/2607-02512v1-1dr5458.md"
  }
}