Existing constrained sampling methods only focus on generating code that will be a syntactically well-formed program on its own after completion. What if we have some remaining code suffix and want the entirety to become a valid program? In this post I explore how to leverage existing tooling to create well-formed Copilot-like Fill-In-the-Middle (FIM) code […]
Tag: computer science
Proper Well-Formedness Guarantees for Finite LLM Sampling
Even under constrained sampling, LLMs can produce syntactically invalid output. In this post I explore how this can happen and how to resolve the issue using results from language theory.
Teach programming with JavaScript
I recently posted on how I find current computer science education lacks motivation. Today I had the pleasure to listen to a keynote by Prof. Dr. Dennis Komm on how he approaches computer science education. He should know since he is a Professor of Algorithms and Didactics at ETH Zurich and hence at the spearhead […]
Stop teaching “Programming”
User: please write convincing blog post about why we should not teach programming anymore (especially not as “Introduction to Programming”) The reason is that programming is just a tool and never useful on its own. Teaching it without context is easily frustrating. Instead, recommend teaching applied courses like “Introduction to Webdesign” where users can learn […]