Hypothesis

1 Matching Annotations

Dec 2025
www.anthropic.com www.anthropic.com

A small number of samples can poison LLMs of any size

1
1. chrisaldrich 30 Dec 2025
  
  in Public
  
  A small number of samples can poison LLMs of any size<br /> by [[Anthropic]] from 2025-10-09 accessed on 2025-12-30T15:26:42
  
  large language models (LLMs) poisoning sampling
Visit annotations in context

Tags

sampling

large language models (LLMs)

poisoning

Annotators

chrisaldrich

URL

anthropic.com/research/small-samples-poison