[Adapt] [Seminar] Representation Engineering: A Top-Down Approach to AI Transparency

[hajin]

Hi, Adapters,Deep neural networks have achieved incredible success across a wide variety of domains, yet their inner workings remain poorly understood. This problem has become increasingly urgent over the past few years due to the rapid advances in large language models (LLMs). Despite the growing deployment of LLMs in areas such as healthcare, education, and social interaction, we know very little about how these models work on the inside and are mostly limited to treating them as black boxes.


In this talk, I will introduce a recent paper titled "Representation Engineering: A Top-Down Approach to AI Transparency". In this work, the authors identify and characterize the emerging area of representation engineering (RepE), which follows an approach of top-down transparency to better understand and control the inner workings of neural networks.




Hope you find this talk useful.




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,

Haoan
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://cs.sjtu.edu.cn/pipermail/adapt/attachments/20231017/febcb39a/attachment.htm>