[Adapt] [Seminar]Approaches on Evaluation ofFactual Precision in Text Generation

[李佳为]

Hi, Adapters,
Nowadays, fact-checking in long text generation remains challenging for validating every piece of information is time-consuming and costly. In this talk, we will explore introduce FActScore, a new evaluation that measures factual precision in long form text generation. We will learn how this scoring system can be an effective tool for evaluating the accuracy of information in various contexts. These insights hold significant implications for the broader field of AI and language modeling. More details can be found via https:// github.com/shmsw25/FActScore.

Hope you all will have fun !

Time: Wed 10 am. - 11:30 am.
Meeting link: https://teams.microsoft.com/l/meetup-join/19%3ameeting_M2VmMTU5MzgtODUzOC00NmU4LTg0MzktNGFjNDdiMmIwYTI1%40thread.v2/0?context=%7b%22Tid%22%3a%225cdc5b43-d7be-4caa-8173-729e3b0a62d9%22%2c%22Oid%22%3a%221a8b9fa0-af57-4a1c-9390-22d1c201d622%22%7d



Best wishes,

Becky