Episode
04/04/25: Constrained Decoding for Code Language Models via Efficient Left and Right Quotienting of Context-Sensitive Grammars with Daniel Melcer
- Podcast
- Boston Computation Club
- Published
- Apr 5, 2025
- Duration seconds
- 3513
- Processing state
not_requested
Actions
POST https://stenobird.com/v1/public/podcasts/boston-computation-club-4031660/episodes/04-04-25-constrained-decoding-for-code-language-models-via-efficient-left-and-right-quotienting-of-context-sensitive-grammars-with-daniel-melcer/transcription-requests
Idempotently request low-priority transcript generation for this episode.GET https://stenobird.com/podcast/boston-computation-club-4031660/04-04-25-constrained-decoding-for-code-language-models-via-efficient-left-and-right-quotienting-of-context-sensitive-grammars-with-daniel-melcer.md
Read the agent-friendly Markdown representation of this episode resource.
Summary
Today Daniel Melcer joined us to present Constrained Decoding for Code Language Models via Efficient Left and Right Quotienting of Context-Sensitive Grammars (https://arxiv.org/pdf/2402.17988). This is work he completed while at Amazon, and it's a really interesting project around how to constrain, guide, and check language models such that they generate valid code within a given context. We really appreciate that Daniel took the time to talk to us and hope you like the talk as much as we did!