Wasi Ahmad Profile picture
Applied Scientist, AWS AI Labs
May 4, 2023 4 tweets 2 min read
#ACL2023NLP
Do causal language models (CLMs) yield good representations with good isotropy and discrimination?

The answer is not always! To address the issue, our ACL2023 paper (arxiv.org/pdf/2210.01185…) proposes ContraCLM.

Joint work with @DejiaoZhang @nihalj_ We show that CodeGen (350M to 16B) pretrained on source code, and text-based CLMs (smaller than GPT2-Large) generated representations suffer from anisotropy and poor discrimination. Image