LLMs have opened up new avenues in NLP with their possible applications, but evaluating their output introduces a new set of challenges. In this talk, we discuss these challenges and our approaches to measuring the model output quality. We will talk about the existing evaluation methods and their pros and cons and then take a closer look at their application in a practical case study.
talk-data.com
A
Speaker
Ada Melentyeva
1
talks
Computational Linguist
Grammarly
Ada Melentyeva — Computational linguist at Grammarly. Worked on inclusive language, fluency, and injecting organizational knowledge into Grammarly suggestions and correctness. Works on a library for metric-based evaluation of prompt output.
Bio from: The Evolution of Prompt Engineering Tooling at Grammarly
Filtering by:
Ensuring the Quality of LLM Output at Grammarly: An Overview and Case Study
×
Filter by Event / Source
Talks & appearances
Showing 1 of 2 activities