Missing important refs in the paper.

jxmorris12 / vec2text

utilities for decoding deep representations (like sentence embeddings) back to text

Other

649 stars 75 forks source link

Missing important refs in the paper. #9

Closed MorinoCheng closed 7 months ago

MorinoCheng commented 8 months ago

Hello John! Congratulations on your nice work that inverts OpenAI embeddings to texts. I just would like you to notice that there is a missing ref in your manuscript. The work is probably one of the earliest works which study the privacy risks of LLM embeddings (also cited by [Song et al.; 2020]).

Pan X, Zhang M, Ji S, et al. Privacy risks of general-purpose language models[C]//2020 IEEE Symposium on Security and Privacy (SP). IEEE, 2020: 1314-1331.

Hope the information can help.

Morino

jxmorris12 commented 7 months ago

Hi, thanks for letting me know. I don't know if I'd say the paper is definitively a 'missing' reference – I think there are many papers at about this level of relevance – but it certainly is interesting and related and I'm surprised I hadn't seen it before. It's great work. Thanks for sharing; I'll keep it in mind for future work or if I ever update the manuscript for the current paper.