SLICE: Semantic Language-Indexed Code Extraction at NeurIPS 2025

Created on December 23, 2025

2025   ·   NeurIPS   SLICE   code-extraction   LatinX   research   ·   research

Our paper SLICE: Semantic Language-Indexed Code Extraction was presented at the LatinX in AI Workshop at NeurIPS 2025 in San Diego.

SLICE presentation at NeurIPS 2025 LatinX in AI Workshop

About SLICE

SLICE introduces a novel approach to code extraction that leverages semantic understanding of natural language queries to identify and extract relevant code segments. The system:

  • Indexes code using semantic representations
  • Matches natural language queries to code functionality
  • Extracts relevant code snippets with high precision
  • Supports multiple programming languages

Research Team

Congratulations to the team on this achievement:

  • Oscar Arámbula
  • Oscar Beltran
  • Omar Guzmán
  • Victor Terrón

Presenting at NeurIPS, one of the premier conferences in machine learning, and being part of the LatinX in AI community was an incredible honor.

Originally shared on LinkedIn



Enjoy Reading This Article?

Here are some more articles you might like to read next:

  • France 24: Discussion on AI Safety and the Future of AI Development
  • IFE Conference: The Future of AI in Education
  • MICAI 2025 Panel: Generative AI's Impact on Education
  • Beyond SWE-Bench: Multi-language Automated Program Repair at MICAI 2025
  • CIMPS 2025: LLM Optimization Techniques for Compliance Assessment