With a few selected prompts, I can identify which LLM engine is being used by #ChatbotArena. At this point, if I am a malicious agent, I can carry out several rounds of evaluation, and each time I identify my target LLM, I can vote accordingly.
Netflix algorithms vs. GPT4. I know I am shooting myself in the foot here a bit, but I am getting much better Netflix recommendations from GPT4 than from Netflix.
I am thrilled to announce that our paper "Off-Policy Evaluation with Deficient Support Using Side Information" has been accepted at #NeurIPS2022 !🥳
@Maurizio_fd M. Restelli @crmpla67
Have you ever noticed how Chinese and American researchers both publish at #NeurIPS, go to the same conference---and then barely cite or talk to each other? @BingchenZhao@gu_yuling@in4dmatics & I have, and we'll be presenting on it at @AiCultures! https://t.co/w0U9zud3dT
We’re thrilled to announce that our @Cleafy LABS research “TeaBot: a new Android malware emerged in Italy, targets banks in Europe” has been quoted into @Apple’s paper “Building a Trusted Ecosystem for Millions of Apps”.
Read the full report➡️https://t.co/iq2deoWYd8
#teabot
I have a vision, a vision in which quantum computing will push artificial intelligence far beyond the limits of our imagination. I believe we are on the brink of a new era.
A collaboration between @moviri, @contentwisetv and @AWS bears quantum fruit! Our research team, led by Paolo Cremonesi, develops an innovative quantum approach to improve recommendation algorithms using Amazon Braket.
Read the AWS Quantum Computing Blog: https://t.co/kB8wCZkVbu
With this first project we are just peeking at what the future promises for all of us, and I hope this is just the beginning of a new, great adventure.
Call me naïve, but that's what makes me a researcher.
I am glad we were invited to present our TOIS paper "A Troubling Analysis of Reproducibility and Progress in Recommender Systems Research" at #SIGIR2021. The second session will be at 2-3 AM for me, I have never given a presentation this late. Let's see how it goes😂
1/ In this episode of #TheDataExchangePod I speak w/ @crmpla67@maurizio_fd of the #RecSys group @polimi. We discussed their recent survey papers on the use of #DeepLearning in recommendation systems: they found it difficult to reproduce research results
https://t.co/FtOPnB6hSr