Past Recording
Share
Star()
Persistent Anti-Muslim Bias in Large Language Models
Thursday Mar 18 2021 23:30 GMT
Please to join the live chat.
Persistent Anti-Muslim Bias in Large Language Models
Why This Is Interesting

With the accelerated development of state-of-the-art large language models and various successful applications. The risk of exploiting and reinforcing the various types of biases using these language models has increased. Identifying and quantifying bias is an important step in developing mitigation strategies. In a recent paper, the authors demonstrated that GPT-3, a state-of-the-art contextual language model, captures persistent Muslim-violence bias. It is possible to reduce bias to a certain extent by introducing words and phrases into the context that provide strong positive associations.

Time of Recording: Thursday Mar 18 2021 23:30 GMT