LLM-Attacker

Mentions · 7d0
Mentions · 30d1
All mentions1
First seenOct 14, 2026
Last mentioned28d ago

Timeline

1 articles · newest first

LLM-Attacker standardizes privacy tests for split language models

In a GPT-2 test, returned gradients raised token recovery from 94.20% to 97.38% and exact-document recovery from 13.71% to 37.77%.

Hugging Face daily papersOct 14, 20:00 UTC · 28d agoAI safety & security1 · 1 read

Appears with

GPT-2ai model · 1

Entities are extracted by the model from each article. First seen Oct 14, 2026, last mentioned 28d ago. Watching keeps it in this browser only (no account); the watchlist page and dashboard alerts use it.