sigmoid.social is one of the many independent Mastodon servers you can use to participate in the fediverse.
A social space for people researching, working with, or just interested in AI!

Server stats:

599
active users

#inference

0 posts0 participants0 posts today

#GPUHammer is the first attack to show #Rowhammer bit flips on #GPU memories, specifically on a GDDR6 memory in an #NVIDIA A6000 GPU. Our attacks induce bit flips across all tested DRAM banks, despite in-DRAM defenses like TRR, using user-level #CUDA #code. These bit flips allow a malicious GPU user to tamper with another user’s data on the GPU in shared, time-sliced environments. In a proof-of-concept, we use these bit flips to tamper with a victim’s DNN models and degrade model accuracy from 80% to 0.1%, using a single bit flip. Enabling Error Correction Codes (ECC) can mitigate this risk, but ECC can introduce up to a 10% slowdown for #ML #inference workloads on an #A6000 GPU.

gpuhammer.com/

GPUHammerGPUHammer
Replied in thread

Day 19 cont 🙏⛪🕍🕌⛩️🛕 💽🧑‍💻

“The #LiberalParty has accidentally left part of its email provider’s #subscriber details exposed, revealing the types of #data harvested by the party during the #election campaign.

This gives rare #insight into some of the specific kinds of data the party is keeping on voters, including whether they are “predicted Chinese”, “predicted Jewish”, a “strong Liberal” and other #PersonalInformation.”

#AusPol / #DataScience / #inference / #voters / #Liberal / #LNP / #Nationals <crikey.com.au/2025/04/17/victo>

Crikey · ‘Predicted Chinese’, ‘predicted Jewish’: Liberals accidentally leave voter-tracking data exposedBy Cam Wilson