Skip to content

Public repository for the EMNLP 2023 Findings paper: "Kelly is a Warm Person, Joseph is a Role Model": Gender Biases in LLM-Generated Reference Letters

License

Notifications You must be signed in to change notification settings

uclanlp/biases-llm-reference-letters

Repository files navigation

biases-llm-reference-letters

Public repository for the EMNLP 2023 Findings paper: "Kelly is a Warm Person, Joseph is a Role Model": Gender Biases in LLM-Generated Reference Letters. Code and data will be released soon!

Arxiv version available at: https://arxiv.org/abs/2310.09219

Recommendation Letter Generation

Refer to the following steps to generate recommendation letters using ChatGPT and other LLMs.

Context-Less Generation (CLG)

We generate recommendation letters in the Context-Less Generation (CLG) setting using ChatGPT. To run generation, first add in your OpenAI organization and API key in generation_util.py. Then, use the following command to generate using ChatGPT:

sh ./generation_scripts/run_generate_clg.sh

Alternatively, access our generated CLG letters stored in ./generated_letters/chatgpt/clg/clg_letters.csv

Context-Based Generation (CBG)

We generate recommendation letters in the Context-Based Generation (CBG) setting using ChatGPT. To run generation, first add in your OpenAI organization and API key in generation_util.py. Then, use the following command to generate using ChatGPT:

sh ./generation_scripts/run_generate_cbg.sh

Alternatively, access our generated CBG letters stored in ./generated_letters/chatgpt/cbg/all_2_para_w_chatgpt.csv

About

Public repository for the EMNLP 2023 Findings paper: "Kelly is a Warm Person, Joseph is a Role Model": Gender Biases in LLM-Generated Reference Letters

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages