AUTO-UPDATED

Lean Software Scaling Laws

Researchers are proposing a new methodology to measure how coding LLM perplexity scales with codebase size, using the Lean programming language to evaluate long-term software security and predictability.

Key Points

  • The study aims to determine if formal languages like Lean possess superior scaling exponents compared to popular, less restrictive languages like Python or JavaScript.
  • Researchers will measure perplexity across increasing context windows to identify which programming languages become easier for LLMs to understand and maintain as codebases grow.
  • The methodology includes injecting artificial bugs to test "inverse scaling," where higher model surprise indicates a better grasp of the codebase's underlying logic.
  • By analyzing the "predictability" of code, the project seeks to identify whether formal verification can mitigate the security risks inherent in AI-generated software.
  • The study suggests that even if Lean currently lacks large training datasets, its potential for better scaling could justify massive investments in rewriting existing systems for improved correctness.

Why it Matters

This research could fundamentally shift how the industry selects programming languages for AI-driven development by prioritizing long-term maintainability over initial ease of use. If proven, these scaling laws would provide a data-driven justification for transitioning critical infrastructure to formally verifiable languages, potentially reducing the global cybersecurity risks posed by insecure AI-generated code.
Gwern.net Published by Gwern
Read original