• Chris Lewis continues to push the envelope of LLM-assisted decompilation in his latest post, building on work on how coding agents could be used to decompile Nintendo 64 games . • Decompilation attempts take time and tokens, so the choice of which unmatched functions to work on matters a great deal. • My original approach prioritized functions based on estimated difficulty. • A logistic regression model ranked candidates using features like instruction count and control-flow complexity, and Claude would always attempt the ’easiest’ remaining function. • That worked remarkably well early on, but it eventually ran out of steam. • At the same time, Macabeus was exploring function similarity via text embeddings of assembly instructions, which then allowed querying for nearby functions in the high-dimensional latent space.

Article Summaries:

  • Chris Lewis continues to push the envelope of LLM-assisted decompilation in his latest post, building on work on how coding agents could be used to decompile Nintendo 64 games . Decompilation attempts take time and tokens, so the choice of which unmatched functions to work on matters a great deal. My original approach prioritized functions based on estimated difficulty. A logistic regression model ranked candidates using features like instruction count and control-flow complexity, and Claude would always attempt the ’easiest’ remaining function. That worked remarkably well early on, but it eve

Sources: