Primary sourceAIArticle··4 min read

Steering a LLM has a price: Apple measures what control costs in fluency

A study from Apple and Pompeu Fabra University shows that the most effective control methods are often the ones that damage generated text the most.

Steering a LLM has a price: Apple measures what control costs in fluency
Source : Iuri Macocco, Pau Rodríguez Lopez, Arno Blaas, Luca Zappella, Marco Baroni, Xavier Suau Cuadros · Apple Machine Learning ResearchView original ↗

In brief

Researchers from Apple and the Universitat Pompeu Fabra systematically compared several LLM conditioning methods: activation steering, prompting, and supervised fine-tuning. Their finding: steering, while effective and cheap, often badly degrades fluency, and works far worse on instruction-tuned models than on base models. The study also puts generation quality back at the center of evaluation, where prior literature mostly just measured whether the target concept was injected or removed.

Older than 30 days: the archive

This article is more than 30 days old

A free account opens articles from the last 30 days in full. The full archive is in the Pro plan.

Create my account

Already a member? Log in

Compare plans →

#llm#apple#steering#research#alignment#evaluation
Original source
On the Effectiveness-Fluency Trade-Off in LLM Conditioning: A Systematic Study
Iuri Macocco, Pau Rodríguez Lopez, Arno Blaas, Luca Zappella, Marco Baroni, Xavier Suau Cuadros
Open the article ↗

For you

Put it to work on your sources.

Free: a month of articles and three sources of your own. Pro: the whole archive and your sources, from €8/month.

For your team

The same machine, on your topics.

A space in your colours, your watch angles, your curators. Pilot open to three companies.

Read next