technique

distillation attacks

techniqueactivedistillation-attacks-8a72764c·1 events·first seen 1mo ago

Aliases: distillation attacks

Co-occurring entities

knowledge distillation Interconnects Anthropic

More like this (12)

distillation Generalized Distillation Self-Distillation ensemble distillation poison fountain attack Weak-to-Strong Distillation skill-based attacks backdoor attack distilabel Model Distillation Random Network Distillation knowledge distillation

Recent events (1)

5Interconnects·1mo ago·source ↗

How much does distillation really matter for Chinese LLMs?

This commentary from Interconnects reacts to Anthropic's post on 'distillation attacks,' examining the role of distillation in the development of Chinese large language models. The piece interrogates how much capability transfer via distillation from frontier models actually explains the progress of Chinese LLMs. It situates the discussion within ongoing debates about knowledge distillation as a competitive and security concern.

Frontier Model Releases Open Weights Progress knowledge distillation Interconnects distillation attacks +2 more